Rotavirus vaccines and vaccination in Latin America

Alexandre C. Linhares1 and Joseph S. Bresee2

ABSTRACT Worldwide, rotaviruses account for more than 125 million cases of infantile gastroenteritis and nearly 1 million deaths per year, mainly in developing countries. Rather than other con- trol measures, vaccination is most likely to have a major impact on rotavirus disease incidence. The peak incidence of rotavirus diarrhea occurs between 6 and 24 months of age. In develop- ing countries, however, cases are not uncommon among children younger than 6 months. G serotypes 1 to 4 are responsible for most disease, but there are indications that in Brazil that G type 5 is of emerging epidemiological importance. Both homotypic and heterotypic responses are elicited during natural rotavirus infection, and the immunological response at the intesti- nal mucosal surface is probably the more consistent predictor of clinical immunity. With the primary objective of protecting children against life-threatening dehydrating diar- rhea, many approaches to rotavirus vaccine development have been attempted. One vaccine, the tetravalent rhesus-human reassortant rotavirus vaccine (RRV-TV), was given licensing approval in the United States of America, introduced to the market, and later withdrawn. A number of studies have found better efficacy of RRV-TV in developed countries than in developing ones. Field trials with a 4 104 plaque-forming units (PFU) preparation of RRV- TV have been carried out in two countries in Latin America, Brazil and Peru. Those trials yielded protective efficacy rates against all rotavirus diarrhea ranging from 18% to 35%. Data from a large catchment trial in Venezuela with a higher RRV-TV dose, of 4 105 PFU/dose, indicated an efficacy rate of 48% against all rotavirus diarrhea and 88% against severe ro- tavirus diarrhea. It appears that breast-feeding does not compromise the efficacy of RRV-TV if three doses of the vaccine are administered. Similarly, possible interference of oral poliovirus vaccine with the “take” of the rotavirus vaccine can be overcome by giving three doses of the rotavirus vaccine or by using a higher-titer formulation of it. Wild enteroviruses, however, may cause primary rotavirus vaccine failure in developing countries. Studies in Peru with RRV-TV have shown a trend towards higher vaccine efficacy rates against “pure” (rotavirus- only) diarrheal episodes. Economic analyses made in the United States indicate that a vaccine that costs less than US$ 9 per dose would lead to a net savings in medical costs. To date, how- ever, cost-benefit studies have not been done in developing countries. In the future, it is possi- ble that some Latin American countries might adapt their polio production facilities to the preparation of rotavirus vaccines for human use. A year after RRV-TV was licensed for vaccination of infants in the United States, the oc- currence of intussusception as an adverse event led to the vaccine’s withdrawal from the mar- ket. The implications of that action, particularly for Latin America, will be addressed in this article, including the need to explore alternative rotavirus candidate vaccines, particularly through the conduct of parallel clinical trials in both developed and developing countries.

1 Fundação Nacional de Saúde, Instituto Evandro gas, Virology Section, Av. Almirante Barroso, 492, 2 United States, Centers for Disease Control and Chagas, Virology Section, Belém, Pará, Brasil; Send 66.090-000, Belém, Pará, Brasil; telephone: (091) Prevention, Division of Viral and Rickettsial Dis- correspondence to: Alexandre C. Linhares, Fun- 2114403; fax: (091) 2261284; e-mail: linhares@ eases, Respiratory and Enteric Virus Branch, At- dação Nacional de Saúde, Instituto Evandro Cha- supridad.com.br lanta, Georgia, United States of America.

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 305 Acute gastroenteritis is a leading as rotaviruses, enteric adenoviruses, will briefly discuss available infor- cause of morbidity and mortality in in- astroviruses, and Norwalk and/or mation on rotavirus structure and fancy and childhood throughout the antigenically related agents (15). It has antigenic composition, epidemiology, world, especially in developing coun- been estimated that viral agents ac- and immunity. We will especially em- tries. Each year diarrheal diseases count for a minimum of 30% of docu- phasize the epidemiological features cause 744 million to one billion cases mented cases of infectious diarrhea in of rotavirus infection in Latin Ameri- of diarrhea in Africa, Asia, and Latin the United States of America (16). can countries. America in children under 5 years of Since their first detection by Bishop age. Of these children, it is estimated et al. (17) in Melbourne, Australia, ro- that 2.4–3.3 million of them die per taviruses have been clearly established GENERAL INFORMATION ON year, corresponding to some 6 600 to as the most common cause of severe, ROTAVIRUS INFECTION 9 000 deaths per day (1–3). An addi- dehydrating gastroenteritis in infants tional hazard in developing countries and young children in both industrial- Structure and is that malnutrition makes children ized and developing countries (18, 19). antigenic characterization more prone to severe disease and, con- It has been estimated that rotaviruses versely, repeated episodes of diarrhea each year worldwide are associated The rotaviruses are classified within may lead to malnutrition (1). with more than 125 million cases of in- a specific genus of the family Reo- In Latin America the incidence rates fantile gastroenteritis and cause a min- viridae. The complete virus particle of diarrheal disease among infants and imum of 418 000 to 520 000 deaths, is often described as smooth, mea- young children vary widely between 85% of which occur in low-income sures about 70 nm in diameter, and different regions and localities, accord- countries (20, 21). An effective vaccine is composed of three protein shells ing to prospective community-based could prevent as many as 326 000 of surrounding the genome. This triple- studies (1). In the poorest periurban these deaths each year (20, 21). layered structure possesses cap- communities of Peru and Brazil, the The similar incidence of the illness someres radiating from the inner to average annual number of diarrheal in both industrialized and developing the outer capsid, like the spokes of a episodes per child under 2 years of age countries suggests that the problem wheel (rota, Latin for wheel) (24). In- reaches 10 or more, while the inci- will not be controlled by merely im- complete (or rough) “single-shelled” dence rate may be as low as 0.7 per proving the water supply, sanitation, particles lacking the outer capsid are child per year among infants (0–1 year or hygiene practices, and that an effec- also commonly found in stool samples old) in Costa Rica (4–7). tive vaccine is required (22, 23). De and tissue culture supernatants. The Estimates for mortality indicate that Zoysa and Feachem (22) calculated viral genome, located within the core, the annual number of deaths caused that the administration of an effective consists of double-stranded RNA that by diarrhea among infants less than 1 rotavirus vaccine within the first 6 can be separated by gel electrophore- year of age also varies widely from re- months of life might prevent at least sis into 11 discrete segments, or genes. gion to region, ranging from 4.5 per 16% of the diarrhea deaths that occur Each of these genes encodes a single 1 000 live births to 21.8 per 1 000 (8, 9). among children under 5 years of age. protein, except gene 11, which seems Nevertheless, data that the Pan Amer- These figures fully justify current glo- to encode two distinct polypeptides ican Health Organization (PAHO) bal efforts directed towards the devel- (25). The outermost layer of the double gathered from 21 countries indicate opment of a vaccine targeted for use in capsid is composed of two surface pro- that the overall mortality from diar- early childhood. Rather than prevent- teins, VP7 and VP4. VP7 is a glycopro- rheal disease in Latin America has de- ing rotavirus infection or mild illness, tein coded by gene 7, 8, or 9, and VP4 clined considerably over the past two the primary goal of rotavirus vaccine is a protease-dependent protein coded decades (10–12). Worldwide observa- strategies would be to protect against by gene 4. These two surface proteins tions suggest that several factors may severe diarrhea that might lead to de- independently induce neutralizing an- have contributed to this reduction, hydration and death. tibodies (19, 26, 27). Specific antibody such as instruction in the correct case This review will focus mainly on the response to these proteins has been management of acute watery diarrhea, current status of rotavirus candidate demonstrated after natural rotavirus particularly at health care centers; the vaccines and the prospect of their fu- infection (28) and also following ad- encouragement of breast-feeding; and ture use in public health programs in ministration of candidate vaccines of improvements in measles immuniza- Latin America, including as a result of animal origin (29). tion coverage (1, 13, 14). the introduction of one rotavirus vac- Of particular interest is sero- During these past two decades, the cine to the market and its subsequent response to VP4 following the natural spectrum of enteropathogens known withdrawal following the occurrence infection of humans. That is because to cause diarrhea has increased, pri- of intussusception among some vacci- antibodies to this protein are known marily through the identification of nated infants. To provide background to broadly react across different sero- a number of novel viral agents, such for the main subject of this report we types (30, 31). Cryoelectron micro-

306 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America scopy studies using computer-assisted serotypes 1 through 4 are recognized to arrheal episodes occurring in this age image enhancement have shown that be globally most common. However, group are associated with group A ro- VP4 forms 60 spikelike structures that other serotypes may be common in taviruses (15, 18, 19). Groups B and C, protrude more than 10 nm from the specific countries (15, 19, 40). In gen- which are antigenically and genet- viral surface (32, 33). VP4 is the viral eral, all human subgroup I rotaviruses ically distinct, also infect humans. hemagglutinin and appears to have bear the G serotype 2 specificity, Group B has been associated with a role in viral virulence (15, 34). The whereas subgroup II strains may be- extensive epidemics of diarrheal ill- cleavage of VP4 by pancreatic trypsin long to G serotypes 1, 3, and 4. There nesses, primarily among adults in into VP5* and VP8* subunit proteins are at least 11 distinct VP4 (P) sero- China (46, 47). Group C viruses, on initiates the process of viral penetra- types. Of these 11, 6 of them have been the other hand, are primarily swine tion into host cells (35). identified among the human rota- pathogens and appear to cause infre- VP7 makes up 90% of the smooth viruses: 1, 2, 3, 4, 5, and 11; these in- quent and sporadic diarrheal disease outer capsid surface, which is perfo- clude four subtypes (36, 41). Recent nu- among children in many parts of the rated by 132 channels penetrating the cleotide and amino acid sequence world. One recent study has shown virion and reaching the viral core. The analyses, on the other hand, allowed that group C rotavirus may be asso- inner shell consists of trimerized VP6 the identification of six distinct ro- ciated with extrahepatic biliary atresia (coded by gene 6), which makes up tavirus VP4 genotypes in humans, des- among infants (48), and other studies over 51% of the virion (19). The core ignated as P[4], P[6], P[8], P[9], P[10], have associated the group with food- is composed of VP1, VP2, and VP3 and P[12]. It has been established that borne institutional outbreaks of gas- proteins, coded by genes 1, 2, and 3, human rotavirus of P genotypes 4, 6, 8, troenteritis (49, 50). respectively. In addition to the six and 9 correspond to P serotypes 1B, Fecal-oral transmission is the most structural proteins, five nonstructural 2A, 1A, and 3, respectively (42). likely route of rotavirus spread, but polypeptides (NSP1 through NSP5) Surveys of rotavirus P types in hun- some limited evidence suggests that have been identified as being encoded dreds of fecal specimens from seven respiratory transmission might also by each of the five remaining rotavirus countries revealed that types P1A[8] occur. While an infected, diarrheic genes (36). Interesting recent studies and P1B[4] are the most common person may excrete approximately (37, 38) have provided evidence that among the rotaviruses of human ori- one trillion infectious particles per NSP4 may cause secretory diarrhea, gin (42–44). In the light of the currently milliliter of stool, the infective dose in thus acting in a similar way to that adopted binary system (VP7 and VP4 a child can be as little as 10 particles of the heat-labile enterotoxin of Es- type specificities) for classification of (51). The main site of viral multiplica- cherichia coli. It is likely that NSP4 will rotavirus strains (40), it is known that tion is the mature enterocytes on the be a target for future rotavirus vaccine three of the four epidemiologically im- villi of the upper small intestine. Over strategies. portant G serotypes (1, 3, and 4) pos- a period of 1–2 days infection spreads Three major antigenic specificities sess P type 1A[8] specificity, while G2 from the proximal small bowel to the are assigned to rotaviruses: group, sub- has P1B[4] specificity (42). However, ileum. Histopathological studies in group, and serotype. Both group and in recently established countrywide animal models have shown mild vil- subgroup specificities are mediated surveillance in the United States, lus shortening, crypt hyperplasia, mi- mainly by VP6, which is the major nearly 10% of rotavirus strains were tochondrial swelling, and infiltration inner capsid protein. To date, seven found to bear unusual P- and G-type of the lamina propria with mononu- groups—designated as A, B, C, D, E, F, specificities (45). In summary, al- clear cells (19, 52). The major mecha- and G—have been distinguished in hu- though the total number of distinct G nism for diarrheal disease induction mans, other mammals, and birds. Only and P types is large, the vast majority in rotavirus infection is likely to be de- group A, however, contains rotavirus of them fall into four G types and two creased absorption of salt and water, strains of epidemiological importance, P types. as a result of intestinal damage and and a vaccine should primarily confer replacement of absorptive epithelial protection to infants and young chil- cells by secreting cells from the villous dren against serious illness caused by Epidemiology and pathogenesis crypts (53). An additional possible this group (36). Two subgroups, I and mechanism is related to the depressed II, can be identified within group A, as Globally, group A rotaviruses are levels of disaccharidases, leading to determined by monoclonal antibodies generally recognized as the most carbohydrate malabsorption and os- (39). The classification of rotaviruses in common enteropathogens identified motic diarrhea (15, 54). Finally, recent serotypes is based on both glycopro- among infants and young children studies have shown that diarrhea tein VP7 (“G” types) and protease-sen- hospitalized with dehydrating diar- may occur in the absence of epithelial sitive protein VP4 (“P” types) antigenic rhea. In both developed and develop- lesion demonstrable by light micro- specificities. Although 14 rotavirus G ing countries it is estimated that 12% scopy, thus raising the question as to types have been identified to date, to 71% (median, 34%) of the severe di- whether other mechanisms might be

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 307 involved in the induction of rotavirus tavirus diarrhea are similar in both in- in the range of 6% to 24%, according gastroenteritis (55). In this respect, it is dustrialized and developing countries, to community-based surveys (74–76). worth noting the findings of Ball et al. the disease is often severe and fatal Data from longitudinal community- (37), which provide evidence that among infants and young children liv- based studies indicate that the inci- NSP4—a nonstructural rotavirus pro- ing in the latter regions, where malnu- dence of rotavirus diarrhea may range tein coded by gene 10—may act as an trition may be an additional hazard from 0.15 to 0.8 episodes per child per enterotoxin in a murine model, induc- (19, 23). Regardless of socioeconomic year (1, 22). ing secretory diarrhea. In addition, status or environmental conditions, it Most of the studies in both inpatient and of interest for vaccine strategies, has been reported that, worldwide, and outpatient settings in Latin Amer- these authors have demonstrated that over 90% of children at the end of the ica involve children aged less than 5 antibodies to NSP4 protect against third year of life have acquired ro- years. Depending on the setting and disease. It has recently been hypothe- tavirus antibody (19, 64). Most studies study, 11.4% to 60% of children with sized that rotaviruses evoke intestinal indicate that the peak incidence of ro- acute diarrhea were found to have ro- fluid and electrolyte secretion by acti- tavirus gastroenteritis occurs between tavirus. For each of the countries, the vation of the enteric nervous system 6 and 24 months of age, although in rates were: Argentina, 12.9% to 34% (56). developing countries cases are com- (77, 78); Brazil, 13% to 40% (79–81); It has been reported, from longitudi- mon among children younger than Chile, 11.4% to 40% (82, 83); Costa nal studies conducted in several coun- 6 months (65). Data from numerous Rica, 45.3% to 60% (84, 85); Ecuador, tries, that asymptomatic rotavirus in- longitudinal studies conducted in both 21.1% (86); Mexico, 14% to 32% fection is frequent and accounts for up developed and developing countries (87, 88); and Venezuela, 30% to 50% to 80% of cases (57–59). On the other show that repeated rotavirus infec- (75). Community-based studies have hand, rotavirus illness is preceded by tions, mostly involving different sero- yielded average incidence rates of ro- an incubation period of 24 to 48 hours, types, are common throughout life, tavirus infantile diarrhea that range has an acute onset, and usually lasts 5 particularly during the early years from 0.17 episode/child/year to 0.8 to 7 days. The clinical picture includes (66–68). In general, both older children episode/child/year, as follows: Ar- vomiting, which starts early in the and adults develop asymptomatic ro- gentina, 0.25 (89); Brazil, 0.3 (68); course of disease, as well as fever, wa- tavirus reinfections, mostly as a result Chile, 0.8 (90); Costa Rica, 0.8 (85); tery diarrhea, mild to moderate dehy- of close (e.g., intrafamilial or institu- Guatemala, 0.8 (85); and Venezuela, dration, and abdominal pain. Both tional) contact with an infected infant 0.17 (75). vomiting and fever remit within the (15, 69). Finally, rotavirus infections Data on the epidemiology of ro- first 2 to 3 days of illness, while diar- among neonates are more likely to be tavirus serotypes in Latin America are rhea usually persists for 5 to 8 days. In asymptomatic, possibly because of still scarce. All four epidemiologically general, laboratory tests indicate nor- passively transferred maternal anti- important G types have been identi- mal serum sodium concentration, mild bodies, the occurrence of apparently fied, and several surveys carried out to moderate elevation of blood urea ni- attenuated strains within hospital to date indicate that predominant ro- trogen, and mild metabolic acidosis nurseries (70), or the immaturity of the tavirus serotypes seem to change over (60, 61). Treatment of rotavirus diar- neonatal gut. time. In general, a particular type will rhea is mainly symptomatic and in- According to a recent review by predominate for 1 or 2 years before the cludes oral (or, less frequently, intra- Cook et al. (18) of 34 studies, cover- emergence of a new dominant strain, venous) replacement of fluid and ing 23 countries, the seasonality of perhaps as a result of immune selec- electrolytes in order to correct dehy- rotavirus infections in temperate re- tion (68, 91–97). The occurrence of ro- dration (62). Of public health impor- gions peaks broadly from autumn to tavirus G type 5 has been claimed to tance is the fact that rotavirus diarrhea spring. In contrast, seasonal peaks are be of emerging epidemiological impor- has been associated with the devel- less distinct in the tropic zones, where tance by some authors and might be opment of lactose intolerance, result- rotavirus infections seem to occur considered for inclusion in the compo- ing in a protracted course of diarrhea year-round. sition of future vaccines (98, 99). Few and exacerbating existing protein- Data from investigations around studies have been made on the distri- calorie malnutrition among infants the world indicate that rotaviruses bution of rotavirus P types in Latin and young children living in tropical account for between 11% and 71% American countries. Recently, an ex- regions (23, 63). (average, 33%) of diarrheal episodes tensive survey carried out in several As previously stressed, the burden requiring hospitalization among chil- states of Brazil indicated that the pre- of rotavirus disease in both developed dren younger than 3 years (18), and dominant strains are similar to those and developing countries is stagger- 40% of nosocomial diarrheal episodes most commonly found in other parts ing, particularly among young chil- among pediatric populations (71–73). of the world: P[8]G1, 43%; P[4]G2, dren aged less than 2 years old (20). The proportion of diarrheal cases 12%; and P[8]G3, 6%. Mixed infections While rates of morbidity related to ro- attributable to rotavirus is generally accounted for 21% of the cases, and

308 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America such unusual types as P[8]G5 ac- recent follow-up study involving Mex- factor-alpha) may play a role in the counted for another 12% of the cases ican infants, it was shown that each protection against infection (126). (99, 100). Bearing G1, G3, or G4 speci- new rotavirus infection increases nat- In addition, passive immunity from ficity, the genotype P[6] has been char- ural protection and reduces the sever- transplacental transfer of antibodies acterized from studies involving new- ity of the diarrhea (112). It remains and breast-feeding may play a role in born infants in hospital nurseries in uncertain whether or not serotype protection against rotaviral disease in Venezuela (42). specificity of the immune response is young infants (26, 127, 128). This view related to clinical immunity during is supported by findings from numer- natural infection. Although homotypic ous studies, indicating that infants do Immunity to rotavirus infection responses seem to predominate, nu- not usually develop severe rotavirus merous studies involving humans and illnesses during their first 6 months Although several studies have been other animal models have shown that of life. Although the protective role of made on the immune response to ro- heterotypic responses are commonly breast-feeding is still controversial, it tavirus infections in humans and the elicited, even after primary infection has been demonstrated that mothers other animals, the mechanisms under- (104, 113, 114). On the other hand, the of uninfected neonates possess higher lying protection against rotavirus dis- occurrence of sequential rotavirus ill- levels of rotavirus-specific neutraliz- ease are not yet fully understood. It is nesses involving the same G type, re- ing antibodies (NA) in colostrum and currently accepted that clinical pro- ported by several investigators, sup- milk than do mothers of infected tection may involve local (mucosal) ports the view that natural infections neonates (129). Nonimmunological and systemic antibodies, and/or the produce incomplete protection (110, protective factors present in human cell-mediated immune system (34, 101, 115, 116). milk, such as mucinlike glycoproteins, 102). Although VP6 has been recog- Both primary and secondary infec- have been claimed by some authors to nized as the most immunogenic rota- tions in humans elicit the development inhibit rotavirus infection (130). Of in- viral protein, genetic studies have of antirotavirus antibodies of the IgG, terest are recent studies involving demonstrated that serum antibodies IgM, and IgA classes in serum, saliva, Bangladeshi children, which indicate directed at either VP4 or VP7 are also and intestinal secretions (28, 30, 117). It that exclusive breast-feeding protects able to neutralize rotavirus and protect has been suggested that high levels of against rotavirus infection during the susceptible hosts (34, 103, 104). In ad- serotype-specific serum neutralizing first year of life; however, the infection dition, a number of studies support the antibodies are closely associated with is postponed to the second year of life hypothesis that VP4 may be more ef- protection against relatively severe (131). Another potential mechanism of fective in evoking virus-specific serum disease (109, 118). According to other passive immunity is the transfer of ma- neutralizing antibodies during natural authors, however, the immunologic ternal IgG to the intestinal lumen in infection. On the other hand, antibod- response occurring at the intestinal both humans and other animals, possi- ies against VP6 in secretions are in- mucosal surface (mainly in terms of bly through transcytosis, leading to dicative of the ability of immunoglob- rotavirus-specific secretory IgA) would virus neutralization prior to infection ulin A (IgA) to neutralize virus, thus probably be a more consistent predic- of enterocytes, if antibodies are pres- reflecting mucosal immunity and re- tor of clinical immunity (102, 119, 120). ent in sufficient concentration (102). sistance to reinfection (105, 106). In addition, the often short-lived na- Also in the context of passive immu- Numerous previous studies have ture of protection induced by natural nity, some studies have shown that suggested that natural infection with infection is in line with the short-term prophylactic protection against infec- rotaviruses confers protection against secretory IgA response (121). tion and clinically significant illness in clinically significant disease during re- To date, investigations on the poten- infants can be afforded by oral admin- infection. Some reports indicate that tial role of cellular immunity in rotavi- istration of either milk or colostrum neonates infected within their first 2 ral infection have been made mainly in containing rotavirus-specific antibod- weeks of life are protected against nonhuman models (122, 123). In mice, ies (132, 133). moderate-to-severe disease but not immune cytotoxic T cells may repre- against reinfection (101, 107). Similarly, sent a major component of host de- infants and children are protected fense in the gastrointestinal tract (124). APPROACHES TO ROTAVIRUS against rotavirus disease following This cytotoxic T lymphocyte activity is VACCINE DEVELOPMENT both symptomatic and asymptomatic linked primarily to CD8 (and possibly primary infection (59, 108). The oc- CD4) cells and seems to mediate virus To date, many approaches to the de- currence of symptomatic reinfection, clearance rather than confer complete velopment of rotavirus vaccines have on the other hand, suggests that, fol- immunity to rotavirus reinfection (102, been attempted, not to prevent infec- lowing natural infection, protection 125). In addition, mechanisms leading tion or mild disease but rather to pro- against the disease may be short-lived to the release of antiviral toxins (e.g., tect against life-threatening dehydrat- or incomplete (95, 101, 109–111). In a interferon-gamma and tumor necrosis ing diarrhea (22, 23). Early vaccine

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 309 trials demonstrated that a low propor- Therefore, in a manner similar to that viruses), and one vaccine (RRV MMU tion of infants receiving a monovalent adopted by Edward Jenner 200 years 18006, serotype 3 virus) was of simian vaccine developed a heterotypic anti- ago (the “Jennerian approach”), a re- origin (106, 139, 140). body response (23). As a result, cur- lated live attenuated agent of nonhu- The RIT 4237 was essentially a high rent strategies are directed toward man origin would be expected to act as primate cell culture passage cold- developing multivalent vaccines that immunogen when administered to hu- adapted strain developed from the bear epitopes similar to the circulating mans (36). The first human candidate bovine strain NCDV. The phase I clin- rotavirus serotypes (23, 134). Several vaccine was an attenuated bovine (G ical trials with this candidate vaccine field trials with reassortant multi- type 6) rotavirus strain, the Nebraska first involved adults and later infants valent rotavirus vaccines have been calf diarrhea virus (NCDV) (137). Of as young as 1 month of age; both of completed around the world. Further- particular interest were the early find- them received a high dose (approx- more, because data from several stud- ings showing that in utero inoculation imately 108 50% tissue culture infec- ies—most of them involving nonhu- of NCDV induces cross-protection in tive dose), administered orally. No man models—indicate that mucosal calves against challenge by human adverse clinical effects were noted immunity rather than circulating anti- rotaviruses (138). among the recipients, and virus shed- body plays a major role in reducing To date, three candidate monova- ding was minimal (141–143). Subse- the incidence or severity of infection lent vaccines prepared from nonhu- quently, several efficacy trials were (119, 120, 135), current strategies are man rotaviruses have undergone field carried out in both developed and de- primarily directed toward developing trials. Two of the vaccines have been veloping countries. Placebo-controlled an orally administered live attenuated derived from a bovine source (RIT efficacy trials among Finnish infants rotavirus vaccine (36). In addition, the 4237 and WC3 strains, serotype 6 achieved protection rates of 50% to recent discovery that rotavirus non- structural protein NSP4 may act as a viral enterotoxin in nonhuman models offers new and exciting approaches to- ward the prevention of rotavirus diar- TABLE 1. Origin and related G- and P-types of rotavirus vaccine rheal illness (37, 136). Using inacti- strains studied to date vated vaccines to immunize pregnant women, in an attempt to enhance the Rotavirus vaccine strains Related G- and P-types passive transfer of specific antibodies Nonhuman origin through the placenta and breast milk, Bovine has also been regarded as an alterna- RIT 4237 P[1], G6 tive approach for protecting against WC3 P[5], G6 rotavirus diarrhea (27). Simian Below, we briefly discuss recent RRV MMU 18006 P[3], G3 Animal-human reassortant strains major strategies adopted for the devel- Bovine-human reassortants opment of a rotavirus vaccine, with WI78-8 P[5], G3 emphasis on rotavirus vaccines of both WI79-9 P[5], G1 human and nonhuman origin and the SC2-9 P[5], G2 D A P[5], G2 human nonhuman reassortant ro- 7 WI79-4 P[8], G6 tavirus vaccines (Table 1). In addition, WI61-4 P[8], G6 we highlight the most significant ob- WC3-quadrivalent P[5], G1,2,3 + P[8], G6 servations resulting from field trials Rhesus-human reassortant strains carried out in various parts of the D X RRV P[3], G1 world, focusing on those conducted in DS1 X RRV P[3], G2 RRV-TV P[3], G1,2,3,4 developing countries. Human rotavirus strains M37 P[6], G1 RV3 P[6], G3 Rotavirus vaccines of 116E P[11], G9 I321 P[11], G10 nonhuman origin 89-12 P[8], G1 Nonreplicating rotavirus vaccines Initial attempts to develop a live at- Viral capsids lacking RNA tenuated rotavirus vaccine were based Recombinant proteins expressed in on the fact that most nonhuman and bacterial/viral vectors Synthetic peptides as subunit vaccines human rotavirus strains share major Nucleic acid vaccines common antigens, particularly VP6.

310 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America 60% against all rotavirus diarrhea, and typic booster response, including an small proportion of inoculated infants over 80% protection from clinically increase in neutralizing antibody lev- (160–162). It has been shown that RRV significant diarrhea (144). In later els to all four major human G types elicits secretory intestinal antibody in phase III studies made in the African (108, 149, 151). immunized infants and, as with the countries of the Gambia (145) and Efficacy trials for WC3 were carried use of bovine-origin vaccine strains, Rwanda (146) and on the White River out in both developed and develop- the neutralizing antibody response to Indian Reservation in the United States ing countries, resulting in variable human rotavirus G serotypes is evi- (147) RIT 4237 failed to provide pro- degrees of protection. In an initial dent only in infants who were seropos- tection against rotavirus disease. Fi- double-blind placebo-controlled effi- itive prior to vaccination (160, 163). nally, in a trial carried out in Lima, cacy trial conducted in Philadelphia, Other serological studies with RRV Peru, comparing the protective effect one dose of vaccine induced 100% pro- have indicated that adult vaccinees of one, two, or three vaccine doses, RIT tection against moderate-to-severe develop both a homotypic neutralizing 4237 vaccine was effective in protect- diarrhea associated with rotavirus G (VP7 and VP4) antibody response to ing against G type 1 rotavirus diar- type 1 infection (139). In contrast, sub- the immunizing strain and a hetero- rhea, particularly against severe di- sequent field trials in the Central typic response to other serotypes. In- sease (148). In the Peru study three African Republic, China, and the fants also develop homotypic neutral- doses of the vaccine conferred 40% United States showed protective effi- izing antibody response but, unlike the protection against any rotavirus diar- cacies that ranged from 0% to 50% adults, they show markedly less het- rheal episode and 58%–75% against against moderate-to-severe rotavirus erotypic immunity (29). severe rotaviral illnesses. The efficacy diarrhea (108, 121, 152). Phase III vaccination studies with rate was 90% against severe diarrheal Major efforts were now directed RRV were carried out in several set- episodes due to G type 1 rotavirus. In toward the evaluation of a vaccine tings around the world, yielding vari- spite of the results in Peru, inconsistent strain originally recovered from a able rates of protection among infants results obtained from field trials of RIT young rhesus monkey with acute di- who were naturally challenged by 4237 led to the discontinuation of fur- arrhea, MMU 18006 (generally called rotavirus G type 1 (163–167). Some ther development of this vaccine (121). RRV, for rhesus rotavirus vaccine) of these trials showed that protective Another monovalent rotavirus can- (153). This vaccine was developed as efficacy rates against moderate-to- didate vaccine of bovine source is the an alternative to those of bovine origin severe rotavirus illness may exceed WC3. This vaccine was developed because it shares neutralization spe- those for all rotavirus diarrhea (164, by the Wistar Institute, Philadelphia, cificity with the G type 3 human ro- 168). Of interest is a vaccine trial con- Pennsylvania, United States, after 12 tavirus (154, 155). In addition, RRV is ducted in Caracas, Venezuela, (169), cell-culture passages of a bovine ro- readily cultured in DBS-FRhL 2 cells, a where the prevailing rotavirus type tavirus strain isolated from a calf with semicontinuous diploid fetal rhesus was G3. RRV proved to be 64% ef- diarrhea (149). The surface protein lung cell line developed by the United ficacious against any rotavirus diar- composition of the WC3 strain indi- States Food and Drug Administration rhea, and vaccine efficacy reached cates that, like NCDV, it belongs to (156). For clinical trials the vaccine was 90% against more severe cases. Ukae serotype G6, but it has a P type 5 spe- prepared from virus produced in 16th et al. (170) in Sapporo, Japan, have cificity (150). The vaccine was orally cell culture passage in the monkey found that a booster dose may en- administered and consisted of live diploid cells, and doses tested were hance the heterotypic protection in- virus in high titer (usually ≥ 107.0 either 104.0 or 105.0 PFU per infant. duced by RRV. plaque-forming units (PFU) per dose). The MMU 18006 vaccine was thor- The wide variability of results from No side effects were noted in several oughly tested in phase I and phase II field trials with monotypic vaccines clinical trials in which more than 500 studies carried out in Finland, Swe- has been attributed to their inability to infants received approximately 107 den, the United States, and Venezuela. induce protection against the four epi- PFU of the vaccine (147, 149). On the In general, this vaccine was found to demiologically important G types in other hand, the fecal shedding of virus be both more immunogenic and reac- infants who had not been primed by was found to be lower than 30% togenic in young children than the rotavirus infection before vaccination among vaccine recipients (149). This bovine-origin rotavirus vaccine strains (36). As a result, in a “modified” Jen- vaccine induced high levels of serum (19, 157–159). Mild, transient fever was nerian approach “second-generation” neutralizing antibodies to the homolo- recorded 2 to 4 days after inoculation polyvalent reassortant candidate vac- gous virus WC3, often approaching in 9% to 40% of infants, but reacto- cines were developed in an attempt 100%; however, few seroconversions genicity was diminished in the pres- to broaden the protective spectrum, to the most prevalent human sero- ence of maternal antibody. In addition, mainly against the four predominant types were observed. Infants who had RRV was shed in feces of more than human rotavirus serotypes, G1 to G4. rotavirus serum antibodies prior to 10% of vaccine recipients, and it in- Bearing human G- or P-type speci- vaccination developed a broad hetero- duced gastroenteric symptoms in a ficity, these reassortants have been

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 311 constructed by using simian (RRV) tively, indicating that VP4 is likely to be the G1, G3, G4, and G9 HRV strains and bovine (WC3) strains (150, 157, more immunogenic than VP7. A larger (175). A second reassortant was WI61- 171, 172). clinical efficacy trial administered three 4, which contained gene 4 of WI61, doses of WI79-9 at 107 PFU/dose to a characterized as a P1A[8], G9 strain. total of 325 infants in two cities in the No adverse reactions were noted Animal-human reassortant United States. Done in Rochester and among infants given either reassor- rotavirus vaccines Philadelphia, the trial resulted in pro- tants and, surprisingly, neutralizing tective efficacy rates of 64% and 74%, antibody responses to WC3 were Bovine-human reassortant rotavirus respectively, against all symptomatic again significantly more frequent than vaccines. Several reassortants were rotavirus episodes (174). Taken to- for human rotavirus parent strains constructed with the purpose of com- gether, the data from the two cities in- from which VP4 genes were obtained bining the safety and immunogenicity dicate that the vaccine was associated (150). A further, double-reassortant of the bovine strain WC3 with the with a 65% reduction in all relatively strain was constructed, WI79 (4+9). It antigenic specificities of predominant severe diarrheal episodes, regardless of had genes for VP4 and VP7 of human human rotavirus serotypes. These the etiology. No adverse reactions were P[8], G1 strain WI79 on a WC3 back- strains were generated by incorporat- attributable to the vaccine within 7 ground. However, no improved im- ing human rotavirus genes (usually 8 days following vaccination. munogenicity of this vaccine was or 9) coding for surface proteins of G1, Two WC3-based reassortant strains, observed among infants after the 7 G2, G3, or P1 specificity to a genome SC2-9 and D7A, were developed with administration of two doses of 10 background of WC3. human rotavirus G serotype 2 speci- PFU. In contrast, and again surpris- The first reassortant strain to be clin- ficity (150). The former strain was de- ingly, more recent investigations in- ically evaluated was WI78-8, which signed to contain the gene encoding dicate that the mixture of WI79-9 has a surface protein composition char- the G2 type VP7 from SC2 on a WC3 with WI79-4 reassortant strains is a po- acterized as P[5], G3. WI78-8 contains genome background, while the latter tent immunogen. Two doses of this genes 2 to 5 from WC3, and the re- was constructed to give a genome mixture given to infants induced 97% maining genes come from the human composition involving genes 2, 4, 10, and 71% neutralizing antibody serore- G3 strain WI78. This candidate vaccine and 11 from WC3 and the remaining sponses to WC3 and HRV WI79, re- proved to be safe among infants re- genome of SC2 strain. Seroresponse spectively (150). ceiving 106 PFU per dose, and two rates among infants given two doses at In an attempt to achieve homotypic doses induced seroresponse rates of 107 PFU per dose of either reassortant protection against the major wild hu- 40% against WC3 and of ≥ 54% against reached about 80% to rotavirus of man rotavirus G types, a quadrivalent a homologous WI78-8 strain (150). WC3 or SC2 specificity, or to both. G2- WC3-based vaccine was constructed. A further bovine-human reassortant type seroresponses were slightly more In one dose it incorporated 107 PFU of rotavirus strain constructed was WI79- frequent among infants vaccinated three VP7 reassortants: WI79-9 (G1), 9, which primarily provided protection with SC2-9 reassortant (55%) than SC2-9 (G2), and WI78-8 (G3). The 6 against the G1 serotype, known to be among infants given the D7A strain fourth component was 5.0 10 PFU the predominant type throughout the (32%). of the human VP4 reassortant WI79-4 United States. The WI79-9 genomic An additional, recent approach with (150, 173). A preliminary evaluation of composition included the gene seg- bovine-human reassortant rotavirus the safety and efficacy of this vaccine ment 9 encoding the VP7 protein de- vaccines was the construction of has been made in the United States, in- rived from the human rotavirus WI79 strains in which the genes coding for volving 417 infants from 10 different (a G1 serotype), with remaining genes predominant human rotavirus P types cities (176). Following a three-dose from the bovine rotavirus strain WC3 were added to a WC3 genome. These regimen of the vaccine, no significant (150, 171). This reassortant, at 107 reassortants were produced on the side effects were detected, and protec- PFU/dose, was not associated with assumption that, unlike reassortants tion against all rotavirus diarrheal epi- any adverse reactions in an efficacy based on human rotavirus (HRV) VP7, sodes was 67%. trial among 77 infants aged 2 to 11 HRV VP4 might preferentially induce Other human-bovine rotavirus reas- months in the United States (150, 173). HRV-specific neutralizing antibodies sortant strains have been assembled Of particular interest in this trial was and presumably yield enhanced pro- using bovine rotavirus UK as the the 100% protective efficacy against ro- tection. One of these reassortants, donor strain. The reassortants result tavirus diarrhea caused by either WI79-4, contained gene 4 of WI79 (G1 from incorporation of the VP7 gene, serotypes G1 or G3, after two doses serotype), which bears genotype 8 derived from human rotavirus strains of the WI79-9 rotavirus strain. In ad- serotype 1A specificity. A broad, D, DS1, P, or ST3 (G serotypes 1, 2, 3, dition, neutralizing antibody serore- cross-reactive protective immunity and 4, respectively), into a UK genome sponses to WC3 and WI79 were noted would be expected, since serotype background (172, 173). Recent safety in 97% and 22% of vaccinees, respec- 1A[genotype 8] is shared by most of and immunogenicity studies with

312 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America these reassortant strains have yielded G serotype 1 (D RRV) or 2 (DS1 United States, D RRV and DS1 satisfactory levels of attenuation, RRV) averaged 66% and 38%, respec- RRV failed to provide an adequate safety, infectivity, and immunogenic- tively, in the first and second rotavirus serotype-specific response in infants ity. These results justify further testing epidemic seasons after vaccination vaccinated at 2 months of age. of a candidate vaccine combining (179). For infants showing a serum IgA In an attempt to broaden the pro- these four reassortants (177). An alter- antibody response, the corresponding tective spectrum of the rotavirus vac- native approach for the UK-based vac- figures were 92% and 59%. cine, a tetravalent formulation was cines has been the construction of re- In a placebo-controlled field trial produced. This RRV-TV formulation assortants containing VP4 protein conducted in the United States, in the incorporated human-rhesus reassor- derived from Wa (a G1, 1A[8] strain), city of Rochester, the D RRV G type tants, representing G1, G2, and G4, along with VP7 protein derived from 1-specific reassortant was found to be along with RRV to represent G3 (Fig- DS1, characterized as being G2, 1B[4] mildly reactogenic and immunogenic, ure 1). Phase I studies have indicated (173). This vaccine has the potential inducing serum antibody responses in that the vaccine is generally safe. to protect against the predominant 71% of the vaccinees (167). Its protec- However, a greater number of febrile worldwide human rotavirus G types. tive efficacy against rotavirus-associ- reactions have been recorded in vacci- ated illness among vaccinees over nees after the first dose than have been three rotavirus seasons was 67.3%, and noted in infants given placebo. Signif- Rhesus-human reassortant rotavirus homotypic protection against the pre- icant IgA ELISA and neutralizing vaccine. Rhesus-human reassortant dominant G type 1 reached 72.8%. antibody (NA) seroconversion rates rotavirus strains have a single human Also in the United States the monova- (66%–83%) to RRV have contrasted rotavirus gene specifying the neutral- lent G type 1 vaccine (D RRV) has with the low levels of NA serore- ization viral protein VP7, with the been evaluated, together with the sponse (from less than 20%, to 43%) to remaining 10 genes coming from the tetravalent preparation, RRV-TV. That human G serotypes 1 to 4 (23, 134, 157, parent, rhesus monkey strain. The national multicenter trial involved 169, 182–185). strains were developed at the National 1 278 healthy infants aged 5 to 25 Several trials have evaluated the Institute of Allergy and Infectious weeks. After three doses, which were protective efficacy of RRV-TV. Three Diseases, Bethesda, Maryland, United given at approximately 2, 4, and 6 trials conducted in the United States States. This work involved the coculti- months of age, the vaccine proved to yielded promising results (164, 186, vation of fastidious human viruses be safe and 54% efficacious against 187). In the first trial (164), three with the tissue culture-adapted RRV any rotavirus diarrheal episode (180). doses of RRV-TV at 4 104 PFU/dose strain MMU 18006, and the subse- Another multicenter efficacy trial in evoked 57% protection against any quent selection of single gene substi- the United States concurrently evalu- rotavirus disease (mostly by rota- tution rhesus-human reassortants ated the D RRV and RRV-TV vac- virus G type 1) over a 2-year period. expressing the VP7 from human G cines. However, no consistent corre- For cases of very severe rotavirus serotypes 1, 2, and 4 (172, 178). These lates of immunity against rotavirus gastroenteritis, protection exceeded strains have been evaluated either in- infection or disease were found, based 80%. A small excess of febrile reac- dividually or as a quadrivalent prepa- on either enzyme-linked immunosor- tions was noted after the first dose. ration containing G types 1, 2, and 4 bent assay (ELISA) IgA or serotype- NA seroresponses to G types 1–4 reassortants, along with RRV as the specific neutralizing antibody sero- were low, again contrasting with the component having homology with the conversions (181). high (nearly 90%) NA seroconversion human G serotype 3. The immunogenicity, safety, and rate to RRV. Interestingly, no consis- Field trials with the single serotype protective efficacy of a single dose of tent correlation was noted between rhesus-human vaccines in Finland and the D RRV vaccine and of the G rotavirus antibody responses after the United States have shown protec- serotype 2 reassortant (DS1 RRV) vaccination and protection, as shown tion following a single dose of 104 PFU vaccine were tested in a recent field by the examination of serum samples (167, 179). The protective efficacy was trial in Lima, Peru (160). Neither diar- from a subset of infants enrolled in not necessarily serotype-specific, and it rhea nor other side effects were associ- the trial (181). appeared to last more than a year. In ated with the administration of either These promising results were con- addition, protection was greatest in of the vaccines. Approximately 50% of firmed in a second large efficacy trial infants with a serum IgA antibody the vaccinees in both vaccine groups in the United States, involving 1 278 response to the vaccine, and was pos- developed an IgA ELISA serore- infants from 24 centers throughout the sibly greater for severe diarrheal epi- sponse, but serotype-specific neutral- country (186). Three doses of either a sodes than for mild illness. In Finland, izing antibodies were induced in 4 105 PFU/dose formulation of where G serotype 1 was prevalent, the fewer than 20% of the participants. In RRV-TV or placebo were given to rates of protection conferred by the rhe- addition, and contrasting with results healthy infants at approximately 2, 4, sus-human vaccine with specificity for from trials in both Finland and the and 6 months of age. With the excep-

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 313 FIGURE 1. Schematic representation of genetic reassortment that occurs during the pro- The protective efficacy of RRV-TV duction of tetravalent rhesus-human reassortant rotavirus vaccine (RRV-TV). Adapted from has been thoroughly evaluated in Kapikian et al. (19), with permission other countries, including Brazil, Fin- land, Peru, and Venezuela. In a recent placebo-controlled efficacy trial, 2 398 Finnish infants received three doses of either the RRV-TV vaccine (4 105 PFU/dose) or placebo, between 2 and Rhesus rotavirus strain, RRV 7 months of age (189, 190). These au- (Serotype G3) thors detected a febrile reaction rate of about 30% among the infants given the vaccine. RRV-TV was found to reduce severe rotavirus disease by 90% and all severe gastroenteritis in young chil- dren by 60%. The first rotavirus vaccine field trial conducted in a developing country, to Human G1 strain Human G2 strain Human G4 strain assess the protective efficacy, im- munogenicity, and safety of RRV-TV, was carried out in Lima, Peru, by Lanata et al. (166). At the ages of 2, 3, and 4 months, 700 infants were given a dose (4 104 PFU) of RRV-TV, an initial dose of vaccine followed by placebo at 3 and 4 months, or a dose of placebo. Neither one nor three doses of RRV-TV was significantly effica- cious against all episodes of rotavirus diarrhea, when compared with the Reassortant, Reassortant, RRV, Reassortant, placebo. However, a three-dose regi- G1 rotavirus strain G2 rotavirus strain G3 rotavirus strain G4 rotavirus strain men of RRV-TV provided moderate but significant protection (35%–66%) against more severe episodes of ro- tavirus diarrhea, mainly those cases Ð Cocultivation in cell culture associated with serotype G1. The inci- Ð Selective antibody pressure against VP7 of RRV dence of diarrhea, vomiting, and fever during 6 consecutive days after the first, second, and third doses did not differ significantly between the vacci- nees and the infants given a placebo. IgA seroresponse after three doses, as tion of a limited febrile reaction after sive trials in the United States (164, determined by ELISA, was 75% in the the first dose, the vaccine was well 186) resulted from the demonstration vaccine group and 24% among the tolerated. The NA seroconversion rates by Pichichero et al. (188) that, although placebo recipients (P < 0.001). On the were also low for G1-4 human sero- it is safe, a single dose of the quadriva- other hand, neutralizing antibody re- types (4% to 31%), whereas serore- lent vaccine at 104 and 105 PFU/dose is sponse to at least one G type, as mea- sponse to RRV among the vaccinees not adequately immunogenic. sured by plaque reduction neutraliza- reached 90%. Of particular interest in The third trial in the United States tion (PRN) assay, was detected among this study were the protective efficacy evaluated the efficacy and safety of the 64% and 12% of the three-dose study rates for RRV-TV: 49% against all high-dose (4 105 PFU/dose) RRV- group and placebo recipients, respec- rotavirus diarrheal episodes, 80% TV with Native American infants aged tively (P < 0.001). against the very severe cases of gas- 6 to 24 weeks old (187). The vaccine In Belém, a city in the northern troenteritis, and 100% against the de- was well tolerated and yielded effi- Brazilian state of Pará, RRV-TV (also hydrating rotavirus illnesses. cacy rates of 50% and 69% against 4 104 PFU/dose) was evaluated for The three-dose regimen schedule for all rotavirus diseases and severe ro- safety, immunogenicity, and efficacy RRV-TV adopted in these two exten- tavirus diseases, respectively. in a 2-year prospective randomized

314 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America TABLE 2. Protective efficacy of rotavirus vaccines in field trials in Latin America generally well tolerated, the RRV-TV induced febrile reactions in 15% of Efficacy (%) vaccinated infants within 6 days after Vaccine/ No. of doses Ages (mo) at All rotavirus Severe rotavirus vaccination, as compared with 7% of a Study site vaccine titers vaccination diarrhea diarrhea those in the control group (P < 0.001). RIT 4237 The vaccine was highly effective Lima, against severe rotavirus gastroenteri- Peru 3 108 2Ð18 40 75 tis. It provided protection rates of RRV 88% against severe diarrhea and of Caracas, Venezuela 1 104 1Ð10 68 100 75% against dehydration and also Lima, produced a 70% reduction in hospital Peru 1 104 3Ð12 29 29 admissions. RRV-TV Lima, Peru 1 (4 104) 2 18 0Ð59 Lima, The issue of intussusception. On 31 Peru 3 (4 104) 2Ð4 24 0Ð47 August 1998 the United States Food Belém, and Drug Administration (FDA) li- 4 Brazil 3 (4 10 ) 1Ð5 35 56 censed the RRV-TV RotaShield® vac- Caracas, Venezuela 3 (4 105) 2Ð4 48 88 cine (Wyeth Laboratories, Inc., Mari- etta, Pennsylvania, United States) for a The titer given is plaque-forming units per dose. vaccination of infants in the United States. In addition, recommendations to use the vaccine among healthy in- fants, in three doses, at 2, 4, and 6 months of age, were made by three double-blind placebo-controlled trial serotypes did not exceed 20% when other organizations in the United involving 540 infants (191) (Table 2). measured by fluorescent focus reduc- States: the Advisory Committee on The infants received three doses either tion assay, but were greater than 40% Immunization Practices (ACIP), the of vaccine or of placebo at 1, 3, and 5 when assayed by PRN. American Academy of Pediatrics, and months of age. The overall vaccine ef- A recent reanalysis of the Peruvian the American Academy of Family ficacy against any rotavirus diarrhea and Brazilian efficacy data indicates Physicians (194). During the following was 35% for the 2 years of surveil- that lower-titer RRV-TV was poten- 11 months, an estimated 1.5 million lance. During the first year there was a tially efficacious against severe rota- vaccine doses were administered to 57% protection against rotavirus diar- virus disease in Peru with one dose of about 900 000 children in the United rheal episodes, most of them associ- vaccine, and against both severe and States. ated with G type 1. In contrast, the very severe rotavirus diarrhea with In July 1999, however, the United vaccine efficacy fell to 12% in the sec- the three doses given in Brazil (192). In States Centers for Disease Control and ond year of follow-up. There was an Brazil the vaccine was 75% protective Prevention (CDC) suspended use of interesting tendency among RRV-TV against very severe diarrheal episodes the RotaShield® vaccine, following the recipients towards enhanced protec- (192). report of 15 cases of intussusception tion against illness associated with an Of particular importance for devel- among vaccine recipients (195). Intus- average of six or more stools per day. oping countries were the results from susception is a condition characterized Low-grade fever occurred on days 3 a large catchment efficacy trial of the by the telescoping of one segment of to 5 after the first dose in 2%–3% of RRV-TV in Caracas, Venezuela (193) intestine into an adjoining segment, the vaccinees. With that exception, no (Table 2). The main objective of the resulting in bowel obstruction. It pri- statistically significant differences in trial was to assess the protective effi- marily occurs among infants. The data the incidence rates of other adverse cacy of high-dose RRV-TV (4 105 on which the CDC made its July 1999 events, such as diarrhea and vomiting PFU/dose) against moderate or severe decision had been gathered from the during the week following vaccina- dehydrating diarrheal episodes. In the Vaccine Adverse Event Reporting tion, were noted between the two study three doses of vaccine or pla- System (VAERS), a passive surveil- study groups. After three doses, about cebo were administered to 2 207 in- lance system operated by the FDA 60% of infants vaccinated with the fants at 4-week intervals, starting at and the CDC. Given the “insensitive” RRV-TV and 33% of those who re- the second month of age. The children character of VAERS, a concern was ceived placebo had IgA seroconver- were then passively followed up to the raised that the true number of intus- sion. On the other hand, NA serocon- age of 24 months in order to detect ro- susception cases might have been versions rates to individual G1 to G4 tavirus diarrheal episodes. Although underestimated.

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 315 By October 1999 a total of 114 possi- difference in intussusception rates was doses of 105 PFU, and the neutralizing ble intussusception cases had been re- not statistically significant (198, 199). antibody seroresponses were specific ported to VAERS, and 99 of them were On the vaccine’s package insert, the predominantly to the vaccine strain confirmed as intussusception. Of these manufacturer had listed this type of rather than to G1 human strains (192). 99, 60 of them occurred within 7 days bowel obstruction as a potential ad- Only one efficacy study has been done of the first, second, or third vaccina- verse reaction, and intussusception with the M37 strain. Using the vaccine tion: 49 of the cases after the first vac- was also discussed in the ACIP recom- with infants 2–6 months old in Fin- cine dose, 10 after the second dose, mendations for use of the vaccine. land, no protection against rotavirus and 1 after the third dose. Surgery was Other forms of prelicensure evidence gastroenteritis caused by G1 viruses required for 32 patients, including were lacking to indicate that intussus- was found (204). bowel resection in 7 of the infants. One ception was associated with vaccina- Another neonatal strain that has fatal case occurred, with a 5-month- tion. In the prelicensure trials there been developed as a vaccine strain is old child. had been no clustering in time of the RV3, characterized as P[6]G3 (35) and Other data came from a retrospec- intussusception cases among the vac- isolated from a 4-day-old infant born tive multistate case-control study con- cinees. Further, it was found that the in a maternity hospital in Melbourne, ducted by the CDC and from a cohort age distribution of the intussusception Australia (205). Neonates who became study conducted in several large cases among the vaccinees was similar naturally infected with this virus de- health maintenance organizations in to that for a large comparison popula- veloped neutralizing antibodies to G1, the United States. Together, these tion (198). Finally, there was no evi- G3, and G4 that still persisted by the studies yielded an initial estimated dence associating wild-type rotavirus age of 3 months. The infants were pro- vaccine-attributable risk of 1 case of with intussusception (198, 200–202). tected against subsequent severe dis- intussusception in 4 300 vaccinated ease caused by rotaviruses of G2 type infants (196). This estimate has subse- during their first 3 years of life (23). quently been revised to approximately Human rotavirus strains Data from a recently completed trial 1 case of intussusception for every indicate that if an immune response is 10 000 vaccinees (United States, Cen- The first non-Jennerian approach to triggered by the vaccine, the child is ters for Disease Control and Preven- vaccination against rotavirus was the protected; on the other hand, if there is tion, unpublished data). use of human strains isolated in nurs- no evidence of immune response, pro- On 22 October 1999 the ACIP con- eries for newborns. This strategy was tection is poor (23, 196). cluded that among vaccinees there is a based on evidence from studies in The 116E (P[11]G9) and the I321 significantly increased risk of intus- Australia, Sweden, the United King- (P[11]G10) strains, isolated from susception within 1 week of a dose, dom, and Venezuela indicating that newborns in India, have also been and therefore withdrew its recommen- the human rotavirus strains that usu- described as potential candidate vac- dation that RRV-TV be administered ally infect neonates seem to be natu- cines, and they are currently being at 2, 4, and 6 months of age (197). rally attenuated (23, 36, 140, 183) and tested among volunteers (107). The However, the ACIP cautioned that confer protection against subsequent former virus is a single gene reassor- such a decision “may not be applicable severe rotavirus illness for up to 3 tant between a human parent strain to other settings, where the burden of years (101, 107). It is interesting that and the VP4 gene of bovine origin, diseases is substantially higher and the both hybridization and sequence whereas the latter is a bovine-human risks and benefits of rotavirus immu- analyses indicate these strains share a reassortant strain bearing multiple nization could be different.” The with- unique VP4-encoding gene that differs genes from bovine origin. drawal of the only licensed rotavirus from the same regions of the G1, G2, Strain 89–12, an attenuated P[8]G1 vaccine underlined the need to re- G3, and G4 strains associated with se- rotavirus human strain originally iso- assess priority activities, particularly vere illness (36). lated from an infant in Cincinnati, future rotavirus vaccine research in The first human-origin rotavirus Ohio, United States, has recently been developing countries (197). This issue vaccine tested was M37. It is a cell tested for safety, immunogenicity, and will be discussed further later in this culture-adapted strain (passage 29), efficacy (206). Low-grade fever was article. characterized as being P[6]G1, that the only side effect detected, over 90% It bears pointing out that intus- was originally isolated from a neonate of the vaccinees seroresponded to the susception did not emerge as an im- in Venezuela who was asymptomatic- vaccine, and protective efficacy against portant adverse reaction during the ally shedding rotavirus. Phase I tri- rotavirus disease was about 90%. prelicensure trials. Prior to licen- als using doses of 104 and 105 PFU Additional approaches towards the sure, among 10 054 vaccinees, 5 cases showed that M37 was safe and moder- development of human rotavirus vac- (0.05%) occurred. In comparison, ately immunogenic (183, 185, 203, 204). cines have included adaptation of a G1 among the 4 633 placebo recipients The vaccine was found to be immuno- human rotavirus type to grow at 26 °C there was only 1 case (0.022%). That genic in 70% of infants given two (207), as well as construction of reas-

316 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America sortant strains. Recently developed (209). It has also been demonstrated to mice and rabbits (215, 216). Aden- cold-adapted reassortant strains pos- that parenteral vaccination of cows oviruses have also been used as poten- sess VP7 serotypes 1 and 2 specifici- with these empty capsids induces a tial expression vectors, given that the ties. Reassortant viruses are also avail- boost in preexisting neutralizing anti- current candidate adenovirus vac- able that contain VP7 serotype 2 or 3 body titers (92). This approach is not cines, for respiratory infections, are gene derived from strains DS-1 or P, currently being considered for human administered orally in enteric coated respectively, and remaining genes use because the available preparations capsules (218). Attempts to produce from human Wa strain, including the may still contain fully infectious ro- attenuated adenoviruses that might gene that codes for VP4 serotype 1A tavirus particles (23). express relevant, sufficiently immuno- (154, 173). genic antigens are under way (219, 220). Recombinant proteins expressed in Nonreplicating rotavirus vaccines bacterial and viral vectors. Current attempts to express recombinant pro- Synthetic peptides as subunit vac- Although the multivalent, human- teins in a suitable vector have focused cines. This approach is based mainly animal reassortant vaccines were the mainly on VP7 and VP4 because of on the ability of specific synthetic VP7 first to be licensed for widespread clin- their importance in inducing neutral- and VP4 epitopes to induce neutraliz- ical application, several other potential izing antibodies (39). Either full-length ing antibodies (26, 221). VP7 peptides approaches to rotavirus vaccination or partial-length genes coding for VP7 that cover most of the amino acid se- are currently being pursued. Most of from human, bovine, and simian ori- quence (amino acids 44 to 295) have these “next-generation” vaccines in- gin have been expressed in a variety of been synthesized, but no significant clude nonreplicating rotavirus vac- bacterial and viral expression systems, neutralizing antibody response has cines, the development of which has including E. coli, vaccinia virus, bac- been achieved following their inocula- been based on advances in molecular ulovirus, Salmonella typhi 21a, and S. tion into experimental animals (23, 26). biology. Compared with the live atten- typhimurium (173, 210, 211). Proteins A few experiments have been carried uated virus strains, these potential synthesized in bacterial systems, espe- out with synthetic peptides matching vaccines would have several theoreti- cially VP7, have been found to be the amino acid sequences of VP4. It cal advantages. These include safety, poorly immunogenic when inoculated has been shown that a synthetic pep- particularly in the immunocompro- into experimental animals. This could tide, corresponding to sequence 296– mised host; potentially lower manu- be explained by the fact that neutraliz- 313 of human rotavirus, binds to a facturing costs; enhanced stability dur- ing epitopes are known to be depen- neutralizing monoclonal antibody to ing transport; removal of possible dent on conformation, and expressed VP4 (25). On the other hand, VP4 pep- reversion to virulence of the attenu- protein may not reproduce conforma- tide 220–223 was found to prime for ated strain; and reduction or even tional requirements for antigenicity of a neutralizing antibody response after elimination of interference by other native outer capsid VP7 (26, 212). In- the inoculation of mice either with in- vaccines or enteric organisms (23). terestingly, cell surface-expressed VP7 tact SA11 or with human ST3 rotavirus Most of these candidate vaccines were of SA11 strain, using recombinant vac- strain (222). developed for parenteral administra- cinia virus, proved to be highly im- Experiments using BALB/c mice tion, although there is some evidence munogenic for mice. This suggests have shown a significant increase in that nonreplicating rotaviruses given that these new techniques of antigen immunogenicity by using synthetic orally can stimulate the immune sys- presentation may improve subunit peptides corresponding to different tem (208). Various approaches to these vaccines (213, 214). well-characterized determinant re- inactivated rotavirus vaccines are A promising, recent approach has gions of protein VP6 of a bovine ro- briefly described below. been the cloning of rotavirus genes tavirus linked to an influenza virus encoding structural proteins in bac- hemagglutinin (223). ulovirus. These recombinant virus Although they are still promising as Viral capsids lacking RNA. The use systems successfully express proteins next-generation vaccines, there is little of differential centrifugation tech- that possess native conformation, if evidence to date that either expressed niques permits the purification of the compared immunologically and bio- viral proteins or synthetic peptides empty capsids that are commonly pre- chemically with native viral proteins will elicit an effective primary local in- sent in cell cultures infected with ro- (215–217). The coexpression of differ- testinal immune response. However, tavirus. These viral particles lacking ent combinations of structural proteins from experimental studies it appears RNA have been used to immunize and their subsequent self-assembly that peptides may be useful for prim- adult female mice subcutaneously results in the formation of viruslike ing an immune response, either prior prior to parturition, leading to protec- particles that are effective immuno- to immunization with a live attenu- tion of their litters from challenge gens when administered parenterally ated virus vaccine or as a booster in

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 317 antibody levels after the administra- other intestinal viruses. In addition, veloping countries compared to devel- tion of a live vaccine (23, 212). the failure to boost with subsequent oped countries. doses can make short-interval vaccina- Since breast milk contains specific tion useless. Secondary vaccine fail- IgA to rotavirus and nonspecific in- Nucleic acid vaccines. A new ap- ures, on the other hand, may be the hibitors of rotavirus replication, there proach to subunit vaccines is based consequence of a waning immunity has been concern that breast-feeding on the direct inoculation of plasmid in the presence of a natural challenge might adversely affect the immuno- DNAs that code for specific viral pro- with a large infectious dose, or of un- genicity of oral vaccines (227, 228). teins (224). An advantage of this tech- usual rotavirus types that are not cov- That issue has been evaluated in at nique is that immunizing proteins ered by the vaccine (226). least 13 clinical trials around the world expressed by host cells are presented Further studies are needed to deter- (228). To date, most of the trials with to the immune system in their native mine specific reasons for vaccine fail- the bovine vaccine strains, the rhesus forms, thus fully mimicking the nat- ure. Furthermore, future immuniza- rotavirus parent strain, and the rhe- ural infection (224, 225). Experimental tion strategies should be compatible sus-human reassortant monovalent studies using adult BALB/c mice have with the Expanded Program on Immu- serotype 1 and tetravalent rotavirus shown that plasmid DNAs encoding nization (EPI) for each country and vaccines suggest that one dose of for murine rotavirus proteins VP4, perhaps preceded by an evaluation of vaccine yields fewer seroresponses VP6, or VP7 elicit both serum anti- cost-effectiveness. The discussion in among breast-fed infants (227–229). bodies and cytotoxic T lymphocyte the following section will cover cur- However, any differences in serore- response. Sufficient to protect mice rent knowledge of rotavirus vaccine- sponses between those breast-fed and against challenge with 100 ID50 homo- related issues in Latin America, rec- those not breast-fed are overcome fol- typic virus, these immune responses ommendations for further rotavirus lowing the administration of three indicate the potential of DNA vaccina- vaccine research, and possible future doses of RRV-TV (185). However, it tion as a new approach towards the immunization strategies. The recent also appears that breast-feeding does control of rotavirus infection (224). withdrawal of the only licensed rota- not compromise the efficacy of rhesus virus vaccine from the market and rotavirus reassortant vaccines, particu- the resulting need to reassess prior- larly if three doses are administered ROTAVIRUS VACCINE ISSUES ity activities will receive particular (180, 230). IN LATIN AMERICA consideration. In developing countries, very few studies have focused on breast-feeding The available results from rotavirus as a factor that may interfere with ei- vaccine trials around the world clearly Possible suppressive effect of ther vaccine take or protective efficacy. indicate that efficacy rates are gener- maternal antibodies In Lima, Peru, Lanata et al. (148) found ally higher in developed countries no significant differences when com- than they are in less developed coun- One possible explanation for the paring seroconversion rates among tries. The history of vaccine trials in reduced efficacy of the rotavirus vac- breast-fed and nonbreast-fed infants Latin America, particularly Brazil and cine in many trials conducted in devel- who received three doses of RIT 4237. Peru, is similar to that of trials in such oping countries is the suppressive ef- Likewise, among breast-fed and non- other developing countries as the Cen- fect of maternal antibodies transferred breast-fed Turkish infants the rates of tral African Republic, China, the Gam- in breast milk or transplacentally. In IgA seroconversions did not differ fol- bia, Myanmar, and Rwanda. With the most tropical regions adults frequently lowing a single dose of RRV-TV (231). exception of studies in Venezuela, experience rotavirus reinfections, none of the live oral rotavirus vaccines which are usually asymptomatic. It is has performed satisfactorily in the to be expected, therefore, that high lev- Early occurrence of developing world; the experiences in els of antibodies are transferred from natural rotavirus infections Latin America are shown in Table 2. mother to child either as transplacen- A number of reasons may explain tally acquired immunoglobulin or as Besides the potential for interference these comparatively low efficacy rates breast milk immunoglobulin (173). In from high levels of rotavirus antibody in developing countries, many of them this respect, possible improvement in that are passively transferred from related to the epidemiology of ro- vaccination performance in develop- mother to child, the high rates of nat- tavirus disease in those areas (226). ing countries may result from dosing ural infection in early life may account First, it has been suggested that pri- with rotavirus vaccine at an age when for the poor immunogenicity of rota- mary vaccine failures may result from the maternally acquired antibodies virus vaccine in developing countries. the presence of maternal antibody at have declined (226). However, this is It is also likely that natural early infec- the time of vaccination and/or inter- problematic because of the relatively tions may confer some protection from ference by oral poliovirus vaccine and early age of rotavirus infection in de- rotavirus disease among both vaccine

318 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America and placebo recipients. This may ex- three with RRV-TV. Ho et al. (234) potential to cause interference and re- plain, for instance, why the RIT 4237 could find no evidence of significant sult in primary rotavirus vaccine fail- vaccine in Peru, which was given be- interference in infants in the United ure in developing countries. Several tween 2 and 18 months of age, yielded States given OPV and RRV simultane- studies in these nations show that higher efficacy in the youngest chil- ously, and a study with RRV and OPV rates of mixed infections with rota- dren, that is, the ones with lower rates in Pakistan gave inconclusive results viruses and other enteric agents are of natural infection prior to vaccina- because of the small numbers of sub- high, often exceeding 25% of all cases tion (148). The occurrence of infections jects (235). Studies in the United States of rotavirus-related diarrheal episodes with rotavirus early in life in develop- showed no statistically significant dif- (19). Some vaccine trials in Latin ing countries makes it imperative that ferences in antibody responses to ro- America have monitored the occur- vaccine programs ensure early, timely tavirus vaccines among infants receiv- rence of diarrheal pathogens other vaccination of children in these areas, ing OPV and RRV-TV concurrently, as than rotaviruses, with results varying and that those who design vaccine tri- compared with those given OPV alone in accordance with the vaccine used. als take into consideration the local epi- (183, 236, 237). The potential for inter- With RIT 4237 in Peru the efficacy rate demiology of rotavirus. ference between RRV-TV in its 104 among children with mixed infections PFU/dose formulation and OPV has was lower than that found in children been investigated in Bangkok by Mi- with “pure,” rotavirus-only infection Possible interference with rotavirus gasena et al. (238). Like other re- (145). Similarly, in subsequent studies vaccine take by oral poliovirus searchers, they found that while OPV with RRV and RRV-TV in Peru where vaccine and/or other enteric agents is likely to interfere with the take of mixed infections were common, a RRV-TV, this can be overcome either trend was observed for higher vaccine Another cause of primary vaccine by giving three doses or using a efficacy rates against diarrheal epi- failure in developing countries may be higher-titer rotavirus vaccine. sodes in which rotavirus was the only interference by oral poliovirus vaccine To date, only one investigation has pathogen isolated (160, 166). However, (OPV) and other enteric viruses (226). been made comparing the efficacy of in trials conducted in Brazil (191) and Whether or not OPV can be concur- rotavirus vaccine given with and with- Venezuela (75, 193), where the rate rently administered with rotavirus out concurrent OPV (236). These au- of mixed infection was 50% and 30%, vaccines has been considered in stud- thors enrolled 1 278 infants aged 5 to respectively, efficacy rates did not ies around the world. While simulta- 25 weeks, at 24 sites located through- differ between infants with mixed in- neous administration of OPV may in- out the United States. The results fections and those with rotavirus-only terfere with the rotavirus antibody showed no major differences in the infections. response (228), this interference can efficacy rates of RRV-TV (4 105 be overcome by administering three PFU/dose) between children given doses of rotavirus vaccine. For in- OPV and RRV-TV simultaneously and Efficacy rates and natural stance, in Yugoslavia rates of rotavirus those who received RRV-TV alone. challenge by serotypes not covered antibody seroconversions among chil- The overall lack of significant inter- by the vaccines dren given both OPV and RIT 4237 action of OPV and rotavirus vaccine is were significantly lower than those encouraging, but this conclusion is It has been shown that live oral ro- among infants receiving rotavirus vac- based mainly on studies in the United tavirus vaccines of animal origin were cine alone (232). Similarly, studies in States. Therefore, caution should be both safe and efficacious (> 80%) Italy with the RIT 4237 vaccine also exercised in extrapolating these results against severe rotavirus diarrhea in showed lower seroresponses among to developing countries, including some populations, but gave no protec- vaccinees who had also received OPV those in Latin America, where the po- tion in others (38). One reason for this (233). Even so, this interference was tency of OPV varies, OPV vaccination variability, particularly in developing noted only following the first dose of schedule may differ from the schedule countries, may be differences in the rotavirus vaccine. No differences were in the United States, and a high rate of rotavirus serotypes making the chal- noted following the second dose. Fi- poliovirus vaccine failure occurs (239). lenge. Thus, during the early trials nally, in the Gambia, no significant Since immune correlates of protection with rhesus rotavirus vaccine in Cara- differences were noted in antibody remain unidentified, it is currently rec- cas, Venezuela, where G3 was the response to RIT 4237 vaccine among ommended that future trials, particu- most common serotype, up to 86% children concurrently vaccinated with larly in developing countries, address efficacy was obtained (169). In Lima, OPV (145). the potential interference by OPV on Peru, however, where serotype G1 Several rotavirus vaccine immuno- the efficacy of rotavirus vaccine rather was predominant, RRV induced only genicity studies have involved simul- than on its immunogenicity (226, 239). a low level of heterotypic protection, taneous vaccination with OPV, includ- Wild enteric viruses and possibly mainly against serotype G1 (160). ing at least two studies with RRV and other enteropathogens also have the High efficacy rates during some RRV

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 319 trials in developed countries where G1 conducted in northern India, for exam- area showed that one dose of 104 PFU was also most prevalent may have ple, the incidence of rotavirus diarrhea of RRV vaccine, of D RRV vaccine been a result of prior priming of par- among infants aged 3–5 months was or of DS1 RRV vaccine, is not suf- ticipants with a VP7 serotype 1 strain. similar to that for infants 6–11 months ficient to induce a high level of When boosted by RRV, these infants old, 0.61 and 0.62 events/child/year, serotype-specific protection among in- developed a broadened heterotypic respectively (107). This suggests that in fants immunized at 2 months of age immunity, whereas only a homotypic developing countries that vaccination (160). A recent phase III study with response was induced among the early in life, prior to the current routine RRV-TV (104 PFU/dose), also carried naive vaccinees (167, 173). Although immunization schedule for OPV and out in Lima, compared the protective available quadrivalent vaccines may for diphtheria-pertussis-tetanus (DPT) efficacies of one and three doses of this provide serotype-specific immunity vaccine, may be beneficial. For devel- vaccine (166). Both dose regimens against each of the four major ro- oped countries it has been recom- failed to induce a significant level of tavirus serotypes (G1 to G4), monitor- mended that rotavirus vaccines be protection against rotavirus diarrhea. ing of strain variation is needed, even given to children along with the rou- In addition, it is evident from studies before introduction of vaccines. In tine schedule (226). However, several in Belém, Brazil, that the protection Brazil and India, for example, recent factors remain to be examined regard- conferred by three doses of RRV-TV studies have shown that additional ing the optimal age for immunization (4 104 PFU/dose) was substantially serotypes not covered by polyvalent in developing countries. Accumulated less in the second year of follow-up vaccines may be of epidemiological evidence from studies in these nations, (12%) than it had been in the first year importance (43, 99, 239). especially India, suggests that the first (57%). These results suggest that the If quadrivalent vaccines are found to vaccine dose should possibly be given duration of vaccine-induced protec- give an unsatisfactory level of protec- earlier than the OPV/DPT schedule tion could perhaps be improved by tion, one approach to pursue, particu- (107). The introduction of an alterna- providing a fourth, booster dose early larly in developing countries, is the tive schedule for rotavirus vaccination in the second year of life (191). generation of a pentavalent vaccine in- would mean increased costs and logis- corporating an additional strain with tical difficulties, but the administration human rotavirus VP4 specificity (36). to neonates of a first dose together Inclusion of vaccine in the An enhanced immunogenicity of such with BCG vaccine may be worth con- Expanded Program on Immunization a vaccine would be expected, since the sidering. Subsequent doses might be VP4 is known to induce broadly cross- administered with both the OPV and In Latin America the routine Ex- reactive neutralizing antibodies (41). DPT vaccination schedule. However, panded Program on Immunization ac- A recently developed candidate vac- further immunogenicity studies are tivities include four visits to the official cine of human origin, a P[8], G1 strain, needed in Latin America to determine public health units in the first 9 was found to be 90% efficacious the potential for interference of both months of life, and these should be the against rotavirus gastroenteritis of high-level maternal antibodies and opportunities for administering ro- any severity in a small-scale trial breast-feeding. tavirus vaccine. A possible later (e.g., conducted in the United States fourth) dose may therefore represent a (206). However, larger efficacy trials logistical problem for completing the are needed, especially in developing Required number of doses schedule. Future introduction of ro- countries, to determine whether a sig- tavirus vaccine in the developing nificant level of protection would also In light of studies carried out thus world will possibly include its com- be achieved where serotypes other far in Latin America, it seems that the bination with forthcoming vaccines than that of the vaccine, even uncom- number of vaccine doses required to against such other enteropathogens as mon strains, are prevalent. achieve good protection is still contro- Vibrio cholerae, enterotoxigenic E. coli, versial and requires further evalua- and Shigella species (240). tion. In a trial conducted in Lima, Maternal immunization against ro- Appropriate age for vaccination Peru, with RIT 4237, for example, the tavirus has been regarded as an addi- results suggested that the protection tional strategy to enhance protection In general, rotavirus infections oc- against serotype 1 rotavirus diarrhea against illness among infants. This is curring within the first 3 months of life provided by one dose is similar to that achieved either through an increase in are either asymptomatic or lead to a achieved with three doses (148). On the concentration of rotavirus-specific relatively mild illness. However, some the other hand, these same authors IgA in breast milk or through the pas- studies carried out in developing have shown that more than one dose sive prenatal transfer of IgG (227). It is countries have shown that the burden of this vaccine may be required to becoming evident, however, that im- of rotavirus disease in this age group give an adequate protection against munizing mothers against rotavirus may be significant. In a cohort study serotype 2. Further studies in the same is unlikely to be a significant public

320 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America health activity in developing coun- this vaccine in developing countries (232). In this respect, the conduct of ef- tries. Instead, it is currently accepted (226). For example, it was agreed that fectiveness trials rather than efficacy that neonatal vaccination would more large-scale immunization with RRV- trials was recommended to expedite effectively prevent early rotavirus ill- TV in developing countries should be the introduction of new vaccines into ness and might also have a priming ef- preceded by additional studies to eval- developing countries (241). fect for subsequent immunizations. uate issues concerning both immuno- Most of the recommendations sum- genicity and vaccine effectiveness. marized above were made during a Available data indicated that, in gen- World Health Organization (WHO) Vaccine introduction eral, rotavirus vaccines are less im- conference held in Geneva, Switzer- in developing countries munogenic in developing than in de- land, in January 1997 that was entitled veloped countries, and the reasons for the Consensus Workshop: Rotavirus The need for a rotavirus vaccine in this remain to be determined. In Latin Vaccines for the Immunization of Chil- both developed and developing coun- America, studies dealing with dose dren in Developing Countries (226), at tries should be based primarily on and schedule of RRV-TV at 4 105 a time when RRV-TV was approaching data from surveillance studies assess- PFU were available only from Ven- licensure in the United States. As there ing the burden of disease, together ezuela, where this vaccine was shown was only one trial showing that ro- with the epidemiologic features of ro- to be immunogenic and effective tavirus vaccine is effective in a devel- tavirus diarrhea in each setting. A first against severe disease (226). However, oping country, WHO recommended step toward this goal, particularly in participants in field trials with lower- that one or two large “demonstration developing countries, could be the in- dose RRV-TV, conducted in Peru and projects” be conducted in Latin Amer- clusion of hospital-based sentinel sur- Brazil, tended to develop lower im- ican countries where the vaccine was veillance systems to gather prevalence mune responses than did the Venezue- likely to be efficacious. Both Carabobo, data on severe diarrhea. In addition, lan infants (166, 191). Additional phase Venezuela, and Mexico City, Mexico, routine characterization of circulating I and II trials were therefore recom- were identified as sites where such tri- rotavirus strains—mainly G and P mended in order to compare these im- als could be performed, and additional types—would be useful for determin- mune responses with those of studies studies were planned for very poor ing predominant strains prior to a vac- in which the vaccine proved to be areas in Africa and Asia. These lat- cination program, documenting signif- highly immunogenic. Also urgently ter settings would certainly repre- icant changes in the rates of prevalent needed were immunogenicity studies sent a major challenge to the rota- types induced by the vaccine, and po- with RRV-TV in Africa and Asia, virus vaccine (226). Such prelicensure tentially aiding in explaining any poor where no field trials had so far been trials could serve as a basis for cost- program effectiveness that occurs. performed. Subsequent phase III trials effectiveness evaluation in developing Certain circulating rotavirus strains were also planned in developing re- countries, and they would also pro- that are potentially more difficult to gions so as to expedite future vaccine vide data to support future official control, such as human rotavirus introduction. These planned vaccine decisions concerning the funding of genotype P[8],G5 in Brazil, could be trials were especially intended to pro- rotavirus vaccination. identified. Plans have been made for duce information on determinants of Available data estimating the eco- the standardization of methodology immune responses, potential effect of nomic impact of immunization against in strain characterization, to improve interference by concurrent administra- rotavirus gastroenteritis came mainly comparability of strain prevalence tion of OPV and/or breast milk, opti- from analyses involving infant popu- data. According to recent recommen- mal age and schedule of vaccination, lations in the United States (242–244). dations (196, 226), such hospital-based the utility of a newborn dose adminis- One study conducted in the United surveillance is needed to establish tered with BCG, and the effect on vac- States demonstrated that an immu- baseline disease estimates to monitor cine response of such public health nization program in which three doses the effect of vaccine introduction; programs as zinc or vitamin A supple- of RRV-TV were included in the rou- detect and explain potential vaccine mentation (226). tine program of childhood immuniza- failures; develop cost-effectiveness Also regarded as essential for imple- tions would prevent more than one models; detect strains potentially not menting rotavirus vaccine programs million cases of gastroenteritis and covered by vaccination; and convince in developing countries is the demon- 33 000 hospitalizations. If the total vac- policymakers of the importance of ro- stration of vaccine efficacy and ef- cine cost was US$ 30 per dose, such a tavirus vaccination campaigns. fectiveness. While efficacy reflects the program would have direct medical When RRV-TV vaccine was ap- performance of a vaccine under ideal- expenses of approximately US$ 100 proaching licensure for use in the ized trial conditions, vaccine effective- per case of rotavirus prevented but United States, a number of formal rec- ness trials provide an indication of result in a total savings of US$ 250 ommendations had been made con- performance under the natural situa- million to society each year (244). An cerning needed further studies with tions faced by a public health program additional analysis was made in the

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 321 United States based on data collected Rotavirus vaccine need, demand, tavirus vaccine, it would be advanta- during a prospective randomized clin- supply, and quality geous to countries in Latin America ical trial that included three oral doses and elsewhere to make use of this of tetravalent rhesus rotavirus vaccine, The global market for rotavirus vac- established system. Otherwise, in- serotype 1 rhesus rotavirus vaccine, cine corresponds approximately to the troduction of rotavirus vaccine into or placebo (242). These authors con- number of global births annually, developing countries would stress the cluded that society should be willing about 130 million (245). This suggests immunization system and hamper to pay between US$ 11 and US$ 12 for the need for a multiplicity of suppliers, implementation of other high-priority immunization against rotavirus, as as well as long-term planning for vaccine coverage programs. well as an additional amount for the supplies of one or more vaccines that intangible benefits of reduced parental are proven effective in developing inconvenience or anxiety associated countries. The intussusception issue and with this illness in infants. An advan- The potential for local production of the need to move rotavirus vaccine tage of rotavirus vaccine was that it live oral rotavirus vaccine is an issue research in new directions could be given during already sched- that should also be considered among uled EPI routine visits, thus incurring strategies for attaining an adequate Given the withdrawal of the Rota- fewer additional costs than other vac- worldwide supply, while also main- Shield® vaccine and the ongoing ur- cines that require extra visits (245). In taining quality. Since manufacturing gent need for an effective rotavirus addition, since the protective efficacy of rotavirus vaccine requires the use vaccine in developing countries, many of rotavirus vaccines appeared to be of cell cultures, some Latin American of the recommendations made at the much higher against a severe disease countries could adapt their polio pro- 1997 WHO workshop on rotavirus than it is against a mild one, a substan- duction facilities to prepare rotavirus vaccines were reassessed recently in tial impact on the direct medical care vaccines for human use. It is unlikely, another WHO meeting, entitled Fu- costs was to be expected (246). however, that such an adaptation will ture Directions for Rotavirus Vaccine Another issue discussed during the be possible before polio vaccine needs Research in Developing Countries WHO rotavirus workshop in 1997 was diminish as a result of progress to- (196). Among the issues addressed at the perception that mothers in devel- ward eradication (226). that February 2000 meeting were data oping countries would have of the The prospect of local production of on the incidence and risk factors for in- efficacy of the vaccine once it was rotavirus vaccines brings a need for tussusception, as well as attributable introduced (226). In some cases, for such official international bodies as risk with current rotavirus vaccine; fu- example, a child may experience as WHO and the United Nations Chil- ture trial designs and directions for many as 30–50 episodes of diarrhea in dren’s Fund (UNICEF) to advise man- disease burden studies; regulatory and the first 5 years of life, with only one or ufacturers, such as with regulatory supply questions; and ethical aspects two of these due to rotavirus. This is a guidelines or requirements on the use of rotavirus vaccines. situation that is quite different from of appropriate cell substrates (not pri- The ethical issues were controversial that of measles immunization, for ex- mary cell cultures), virus seed banks, and received particular attention. A ample, where children are protected and standardized vaccine testing as- key consideration is that the risks and against a distinctive, easily recognized says (247). An initial source of such benefits of rotavirus vaccination may disease. Therefore, the impact of rota- guidelines could be ones from the differ substantially between developed virus vaccination will not be so evi- United States Food and Drug Admin- and developing countries (196). For ex- dent to parents, but physicians and istration (248). Since they will be deal- ample, each year there may be only 20 other health care workers should see a ing with a live vaccine, it is critical that to 40 deaths from rotavirus gastroen- substantial decline in the number of manufacturers in developing coun- teritis in the United States but 15 000 diarrhea-related clinic visits, hospital- tries obtain all materials used in the to 30 000 in Bangladesh. This contrast izations, and deaths. Therefore, among production process from high-stan- raises the question as to whether it the strategies to establish rotavirus dard sources that assure they have would be ethical or moral not to use vaccination in developing countries been tested for all adventitious agents. RRV-TV in Bangladesh or other coun- should be efforts to correctly alert par- WHO and UNICEF have said that tries where a child’s risk of death from ents to the importance of the disease the “backbone” of the EPI is the cold rotavirus is likely to be much greater (245). This task will be even more dif- chain, that is, the network of refrigera- than the risk of death from vaccine- ficult given the link between rotavirus tors, freezers, and cold boxes orga- associated intussusception. vaccines and the risk of developing in- nized by teams of people throughout Preliminary data from Latin Amer- tussusception. Parents should be made the world (239). If special handling, ica also suggest that the potential ben- fully aware that the benefits of vacci- storage, and shipment requirements efit of RRV-TV vaccine far outweighs nation will largely outweigh any pos- beyond those applied to other routine the risk of intussusception (196). In sible adverse reaction. vaccines are not necessary for ro- Venezuela, the rate of rotavirus hospi-

322 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America talization was found to be 158 times the potential risk for intussusception, rather than protecting children from rate of hospitalization for naturally oc- and they should be conducted in rotavirus infection or mild illness. curring intussusception. A risk-benefit parallel in developed and develop- Although several live attenuated oral study conducted in Peru indicated that ing countries. vaccines have been obtained, the RRV- a rotavirus vaccine would prevent • Further testing of RRV-TV in devel- TV rhesus-human reassortant vaccine 1 440 deaths and 23 000 hospitaliza- oping countries is ethical, given the was the first licensed for human use. tions, as compared to an estimated 78 potential favorable benefit-risk ratio However, the recommendation for use cases of intussusception that might be in these settings. Proper manage- of this vaccine was recently withdrawn related to the vaccine. Although under- ment for possible cases of intussus- in the United States because of the oc- reporting may have occurred for some ception is strongly advised. These currence of intussusception as an ad- regions in Brazil, official data indicate a studies should not be undertaken, verse event from vaccination. rate of hospitalization for intussuscep- however, without the assurance that Available results from several field tion of only 3.5/100 000/ year among RRV-TV would be available for gen- trials with live oral rotavirus vaccines children aged less than 1 year. eral use, if results from the trials indicate that efficacy rates of RRV-TV The participants at the recent WHO prove to be favorable. were generally higher in industrialized rotavirus meeting agreed that, on ethi- • The use of a standardized WHO countries than they were in developing cal grounds, clinical trials with RRV- protocol is recommended for dis- nations, although the recent data from TV should proceed in developing ease burden studies in selected de- Venezuela are encouraging. In trials in countries, provided that a number of veloping countries. South America, and in general in de- conditions and safeguards are met. • Research activities dealing with both veloping countries, primary vaccine Among these would be a favorable the pathogenesis and epidemiology failure may have resulted from high risk-benefit ratio, based on evidence; of intussusception are strongly rec- maternal antibody levels and interfer- obtaining official ethical approvals ommended, particularly epidemi- ence from enteroviruses. Secondary and informed consents that are sensi- ologic studies in countries where vaccine failures may be a consequence tive to local cultures; a willingness by clinical trials with new rotavirus of a waning immunity in the presence the target community to support the vaccines are being planned. of a natural challenge with a large in- trial; and monitoring of adverse events • WHO should provide continuing fectious dose or unusual rotavirus during the trial. support to national regulatory au- types not covered by the vaccine. Although the conduct of further clin- thorities in developing countries so The withdrawal of recommendation ical trials with RRV-TV was regarded that international standards for vac- for use of RRV-TV in the United States as ethical, a number of other concerns cine regulation are achieved. will very likely make acceptance of were raised at the Geneva meeting re- • Laboratory surveillance of rotavirus this vaccine in other developed coun- lating to using the vaccine in develop- strains is especially needed in Africa tries difficult. Although it is conceiv- ing countries (196). The availability of and Asia. able that benefits from rotavirus vacci- the vaccine is uncertain, even if results nation in the developing areas of the of future trials are promising in re- world outweigh the risk of intussus- gards to both efficacy and risk of devel- GENERAL CONCLUSIONS ception, it is likely that original plans oping intussusception. Furthermore, for further studies with RRV-TV in de- problems could be faced as to the po- The global need for an effective ro- veloping countries will be changed litical acceptability of RRV-TV, particu- tavirus vaccine has been made clear substantially. larly given that new candidate rota- from worldwide studies, which indi- The safety of RRV-TV has been as- virus vaccines will shortly be available cate that this virus is the major cause sessed through field trials around the for clinical trials. Investigators at possi- of severe dehydrating diarrhea in world, and mild fever after the first ble sites for RRV-TV testing might find young children. Although morbidity vaccine dose appeared to be the only it difficult to obtain formal clearances rates are similar in developed and de- detectable side effect. Intussusception from local authorities because of the veloping countries, rotaviruses are es- was detected among vaccine recipients intussusception risk identified in the timated to cause as many as one mil- but at rates that did not differ signifi- United States. lion deaths annually among infants cantly from those of placebo recipi- A set of six key conclusions and rec- and young children under 5 years of ents. However, the safety of RRV-TV ommendations came out of the recent age in developing nations. must be reviewed, given the appear- WHO meeting (196). They were: During the past 15 years consider- ance of intussusception following vac- able progress has been made toward cination in the recent large-scale use in • The rapid development of new ro- developing a rotavirus vaccine tar- the United States. tavirus vaccine candidates should geted for use in early infancy. The Studies carried out in India and be encouraged. Trials with such main goal of this vaccine is the preven- other developing countries have candidate vaccines must assess the tion of severe dehydrating diarrhea, shown that the burden of rotavirus

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 323 disease within the first 3 months of life impact of rotavirus vaccine will not be postvaccine intussusception in diverse may be significant. This suggests it so evident to parents as is the case settings during the recent experience may be beneficial to administer the with measles immunization, for exam- with RRV-TV in the United States. first rotavirus vaccine dose together ple. Nevertheless, physicians and Without these data, compelling ar- with BCG, early in life, followed by other health care workers will likely guments for and against use of the vaccinations during the subsequent note a considerable decline in the vaccine in developing countries were routine immunization visit. At the number of clinic visits, hospitaliza- difficult. Testing of new vaccines in time of the first dose, however, mater- tions, and deaths caused by diarrhea. parallel in both developed and devel- nal antibodies may affect immune Prospects for local production of ro- oping countries might have changed response. Since naturally occurring tavirus vaccine must be considered. this situation, and such testing in the intussusception clusters among those Strategies are needed that will lead to future would certainly speed intro- 3 to 6 months old, neonatal rotavirus a worldwide vaccine supply, while duction of effective vaccines into set- vaccination may limit the risk of chil- still maintaining quality. This might tings where the benefit will be the dren developing this clinical condition include adapting polio production fa- greatest. In addition, studies are when receiving later doses. cilities to prepare rotavirus vaccine in needed to better understand both the It is likely that rotavirus vaccines, some Latin American countries. Such pathogenesis and the epidemiology of like all new vaccines, will cost signifi- an adaptation is not practical, how- intussusception in developed and de- cantly more than the vaccines included ever, until there is more progress to- veloping countries. in existing vaccination programs. ward eradication, and polio vaccine Finally, studies on disease burden These costs will represent a major bar- requirements diminish. are increasingly important, using stan- rier to rotavirus immunization in less- In Latin America it would be ad- dardized protocols. There is a need to developed areas, where the disease has vantageous if special handling, stor- develop regional laboratory networks a major public health impact. To over- age, and shipment requirements for to share expertise and resources, as come this problem, WHO and other rotavirus vaccine did not go beyond well as to monitor circulating rota- agencies are encouraging research di- those already established for routine virus strains. This will be particularly rected at reducing dosage schedules, vaccines. important in Brazil and other coun- identifying new methods to fund vac- The intussusception issue under- tries where a marked strain diversity cine purchases, and exploring options scores the need to explore alternative has been observed. for local development and production rotavirus candidate vaccines. In addi- of vaccines in developing countries. tion, it will be important to conduct fu- When rotavirus vaccine is intro- ture trials of new candidate vaccines Acknowledgments. We are most duced in developing countries, strate- concurrently in developed and devel- indebted to Prof. Ralph Lainson for gies should be applied to alert parents oping countries. The risks of adverse his attention in reviewing this manu- to the importance of the disease. Par- reactions to any vaccine, such as intus- script and to Dr. Roger I. Glass for en- ents should be particularly aware of susception following administration of couragement and helpful suggestions. the potential benefits of the rotavirus RRV-TV, may be quite different in in- We also thank John O’Connor, M.S., vaccine, while still knowing that intus- dustrialized and developing countries. and Daniele M. de Barros for editorial susception has been identified as a It is clear that it would have been help- assistance in the preparation of the rare adverse event of vaccination. The ful to have had data regarding risks of manuscript.

REFERENCES

1. Bern C, Martines J, de Zoysa I, Glass RI. 3. Thielman NM, Guerrant RL. From Rwanda Huascar, Peru. Am J Epidemiol 1989;129: The magnitude of the global problem of di- to Wisconsin: the global relevance of diar- 785–799. arrhoeal disease: a ten-year update. Bull rhoeal diseases. J Med Microbiol 1996;44: 5. Lopez de Romana G, Brown KH, Black RE, WHO 1992;70:705–714. 155–156. Kanashiro HC. Longitudinal studies of in- 2. Snyder JD, Merson MH. The magnitude of 4. Black RE, Lopez de Romana G, Brown KH, fectious diseases and physical growth of in- the global problem of acute diarrhoeal dis- Bravo N, Balazar OG, Kanashiro HC. In- fants in Huascar, an underprivileged peri- ease: a review of active surveillance data. cidence and etiology of infantile diar- urban community in Lima, Peru. Am J Bull WHO 1982;60:605–613. rhea and major routes of transmission in Epidemiol 1989;129:769–784.

324 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America 6. Schorling JB, Wanke CA, Schorling SK, national immunization schedules. Health vere diarrhea in infants and young chil- McAuliffe JF, de Souza MA, Guerrant RL. Econ 2000;9:19–35. dren. Trends Microbiol 1994;2:242–249. A prospective study of persistent diarrhea 22. De Zoysa I, Feachem RV. Interventions for 37. Ball JM, Tian P, Zeng CQ-Y, Morris AP, among children in urban Brazilian slums. the control of diarrhoeal diseases among Estes MK. Age-dependent diarrhea in- Patterns of occurrence and etiologic agents. young children: rotavirus and cholera im- duced by a rotaviral nonstructural glyco- Am J Epidemiol 1990;132:144–156. munization. Bull WHO 1985;63:569–583. protein. Science 1996;272:101–104. 7. Simhon A, Mata L, Vives M, Rivera L, Var- 23. Bishop RF. Development of candidate ro- 38. Glass RI, Lang DR, Ivanoff BN, Compans gas S, Ramirez G, et al. Low endemicity and tavirus-vaccines. Vaccine 1993;11:247–254. W. Introduction: rotavirus-from basic re- low pathogenicity of rotaviruses among 24. Ludert JE, Gil F, Liprandi F, Esparza J. The search to a vaccine. J Infect Dis 1996 Sep;174 rural children in Costa Rica. J Infect Dis structure of the rotavirus inner capsid stud- Suppl 1:S1–S2. 1985;152:1134–1142. ied by electron microscopy of chemically 39. Greenberg HB, Valdesuso J, van Wyke K, 8. Barros FC, Victora CG, Vaughan JP, Teix- disrupted particles. J Gen Virol 1986;67: Midthun K, Walsh M, McAuliffe V, et al. eira AM, Ashworth A. Infant mortality in 1721–1725. Production and preliminary characteriza- southern Brazil: a population based study 25. Brown DW, Mathan MM, Mathew M, tion of monoclonal antibodies directed at of causes of death. Arch Dis Child 1987;62: Martin R, Beards GM, Mathan VI. Rota- two surface proteins of rhesus rotavirus. J 487–490. virus epidemiology in Vellore, South In- Virol 1983;47:267–275. 9. Bailey P, Tsui AO, Janowitz B, Dominik R, dia: group, subgroup, serotype, and elec- 40. Estes MK, Cohen J. Rotavirus gene struc- Araujo L. A study of infant mortality and trophoretype. J Clin Microbiol 1988;26: ture and function. Microbiol Ver 1989;53: causes of death in a rural north-east Bra- 2410–2414. 410–449. zilian community. J Biosoc Sci 1990;22: 26. Matsui SM, Offit PA, Vo PT, Mackow ER, 41. Gorziglia M, Larralde G, Kapikian AZ, 349–363. Benfield DA, Shaw RD, et al. Passive pro- Chanock RM. Antigenic relationships 10. De Borrero M. Program for diarrheal dis- tection against rotavirus-induced diarrhea among human rotaviruses as determined ease control in Cali, Colombia. Bull PAHO by monoclonal antibodies to the hetero- by outer capsid protein VP4. Proc Ntl Acad 1984;18:86–89. typic neutralization domain of VP7 and the Sci USA 1990;87:7155–7159. 11. Pan American Health Organization. Mor- VP8 fragment of VP4. J Clin Microbiol 42. Gentsch JR, Woods PA, Ramachandran M, tality due to intestinal infectious diseases 1989;27:780–782. Das BK, Leite JP, Alfieri A, et al. Review of in Latin America and the Caribbean, 27. Molineaux PJ. Human immunity to ro- G and P typing results from a global collec- 1965–1990. Epidemiol Bull 1991;12(3):1–10. tavirus. J Med Microbiol 1995;43:397–404. tion of rotavirus strains: implications for 12. World Health Organization. Diarrhoeal 28. Richardson SC, Bishop RF. Homotypic vaccine development. J Infect Dis 1996;174 diseases control programme: an overview serum antibody responses to rotavirus pro- Suppl 1:S30–S36. of the problem in the Americas. Wkly Epi- teins following primary infection of young 43. Gunasena S, Nakagomi O, Isegawa Y, Kaga demiol Rec 1982;57:353–355. children with serotype 1 rotavirus. J Clin E, Nakagomi T, Steele AD, et al. Relative 13. Feachem RG. Interventions for the control Microbiol 1990;28:1891–1897. frequency of VP4 gene alleles among of diarrhoeal diseases among young chil- 29. Green KY, Taniguchi K, Mackow ER, human rotaviruses recovered over a 10- dren: supplementary feeding programmes. Kapikian AZ. Homotypic and heterotypic year period (1982–1991) from Japanese chil- Bull WHO 1983;61:967–979. epitope-specific antibody response in adult dren with diarrhea. J Clin Microbiol 1983; 14. De Zoysa I, Rea M, Martines J. Why pro- and infant rotavirus vaccinees: implications 31:2195–2197. mote breast-feeding in diarrhoeal disease for vaccine development. J Infect Dis 1990; 44. Steele AD, Garcia D, Sears J, Gerna G, control programmes. Health Pol Plan 1991; 161:667–679. Nakagomi O, Flores J. Distribution of VP4 6:371–379. 30. Grimwood K, Lund JCS, Coulson BS, Hud- gene alleles among human rotaviruses by 15. Blacklow NR, Greenberg HB. Viral gas- son IL, Bishop RF, Bares GL. Comparison of using probes to the hyperdivergent gene of troenteritis. N Engl J Med 1991;325: serum and mucosal antibody responses fol- the VP4 gene. J Clin Microbiol 1993:31: 252–264. lowing severe acute rotavirus gastroenteri- 1735–1740. 16. Kotloff KL, Wasserman SS, Steciak JY, Tall tis in young children. J Clin Microbiol 1988; 45. Ramachandran M, Gentsch JR, Parashar BD, Losonsky GA, Nair P, et al. Acute diar- 26:732–738. UD, Jin S, Woods PA, Holmes JL, et al. De- rhea in Baltimore children attending an 31. Ward RL, Knowlton DR, Schiff GM, tection and characterization of novel ro- outpatient clinic. Pediatr Infect Dis J 1988; Hoshino Y, Greenberg HB. Relative con- tavirus strains in the United States. J Clin 7:753–759. centrations of serum neutralizing antibody Microbiol 1998:36:3223–3229. 17. Bishop RF, Davidson GP, Holmes IH, Ruck to VP3 and VP7 proteins in adults infected 46. Hung T, Chen G, Wang C, Chou Z, Chao T, BJ. Virus particles in epithelial cells of duo- with a human rotavirus. J Virol 1998;62: Ye W, et al. Rotavirus-like agent in adult denal mucosa from children with viral gas- 1543–1549. non-bacterial diarrhoea in China [letter]. troenteritis. Lancet 1973;2(7841):1281–1283. 32. Prasad BV, Burns JW, Marietta E, Estes MK, Lancet 1983;2(8358):1078–1079. 18. Cook SM, Glass RI, LeBaron CW, Ho MS. Chiu W. Localization of VP4 neutralization 47. Hung T, Chen G, Wang C, Yao H, Fang Z, Global seasonality of rotavirus infections. sites in rotavirus by three-dimensional Chao T, et al. Waterborne outbreak of Bull WHO 1990;68:171–177. cryo-electron microscopy. Nature 1990;343: rotavirus diarrhoea in adults in China 19. Kapikian AZ, Chanock RM. Rotaviruses. 476–479. caused by a novel rotavirus. Lancet 1984; In: Fields BN, Knipe DM, Chanock RM, 33. Yeager M, Dryden KA, Olson NH, Green- 1(8387):1139–1142. Hirsch MS, Melnick JL, Monath TP, et al., berg HB, Baker TS. Three dimensional 48. Riepenhoff-Talty M, Gouvea V, Evans MJ, eds. Virology. 3rd. ed. Philadelphia: structure of rhesus rotavirus by cryoelec- Svensson L, Hoffenberg E, Sokol RJ, et al. Lippincott-Raven Publishers; 1996. pp. tron microscopy and image reconstruction. Detection of group C rotavirus in infants 1657–1708. J Cell Biol 1990;110:2133–2144. with extrahepatic biliary atresia. J Infect Dis 20. United States, Institute of Medicine. The 34. Offit PA, Shaw RD, Greenberg HB. Passive 1996;174:8–15. prospects for immunizing against rota- protection against rotavirus-induced diar- 49. Mackow ER. Group B and C rotaviruses. In: virus. In: IOM, ed. New vaccine develop- rhea by monoclonal antibodies to surface Blaser MJ, Smith PD, Ravdin JI, Greenberg ment: establishing priorities. Volume 2: proteins VP3 and VP7. J Virol 1986;58: HB, Guerrant RL, eds. Infections of the gas- Diseases of importance in developing 700–703. trointestinal tract. New York: Raven Press; countries. Washington, D.C.: National 35. Konno T, Suzuki H, Kitaoka S, Sato T, 1995. p. 983–1008. Academy Press; 1986. pp. 308–318. Fukuhara N, Yoshie O, et al. Proteolytic en- 50. LeBaron CW, Furutan NP, Lew JF, Allen JR, 21. Miller MA, McCann L. Policy analysis of hancement of human rotavirus infectivity. Gouvea V, Moe C, et al. Viral agents of gas- the use of hepatitis B, Haemophilus in- Clin Infect Dis 1993;16:92–97. troenteritis. Public health importance and fluenzae type B-, Streptococcus pneumo- 36. Hoshino Y, Kapikian AZ. Rotavirus vac- outbreak management. MMWR 1990;39: niae-conjugate and rotavirus vaccines in cine development for the prevention of se- 1–24.

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 325 51. Ward RL, Bernstein DI, Young EC, Sher- 65. Huilan S, Zhen LG, Mathan MM, Mathew tis associated with rotavirus presence in wood JR, Knowlton DR, Schiff GM. Human MM, Olarte J, Espejo R, et al. Etiology of Brazilian children: evaluations on preva- rotavirus studies in volunteers: determina- acute diarrhoea among children in devel- lence, therapy and age group. J Diarrhoeal tion of infectious dose and serological re- oping countries: a multicentre study in five Dis Res 1985 Jun;3(2):78–83 sponse to infection. J Infect Dis 1986;154: countries. Bull WHO 1991;69:549–555. 80. Houly CAP, Uchoa MMM, Zaidan AME, 871–880. 66. Bernstein DI, Sander DS, Smith VE, Schiff Gomes-Neto A, de Oliveira FM, Athayde 52. Oishi I, Kimura T, Murakami T, Haruki K, GM, Ward RL. Protection from rotavirus MAG, et al. Electrophoretic study of the Seto Y, Minekawa Y, et al. Serial observa- infection: 2-year prospective study. J Infect genome of human rotavirus from Maceió, tions of chronic rotavirus infection in an Dis 1991;164:277–283. Brazil. Brazilian J Med Biol Res 1986;19: immunodeficient child. Microbiol Immunol 67. Cravioto A, Reyes RE, Trujillo F, Uribe F, 33–37. 1991;35:953–961. Navarro A, De La Roca JM, et al. Risk of di- 81. Linhares AC, Monção HC, Gabbay YB, 53. Mavromichalis J, Evans N, McNeish AS, arrhoea during the first year of life associ- Araujo VLC de, Serruya AC, Loureiro ECB. Bryden AS, Davies HA, Flewett TH. Intesti- ated with initial and subsequent coloniza- Acute diarrhoea associated with rotavirus nal damage in rotavirus and adenovirus tion by specific enteropathogens. Am J among children living in Belém, Brazil. gastroenteritis assessed by d-xylose malab- Epidemiol 1990;131:886–904. Trans R Soc Trop Med Hyg 1983;77: sorption. Arch Dis Child 1977;52:589–591. 68. Linhares AC, Gabbay YB, Freitas RB, 384–390. 54. Kerzner B, Kelly MH, Gall DG, Butler DG, Travassos da Rosa ES, Mascarenhas JDP, 82. Avendãno LF, Calderon A, Macaya J, Pren- Hamilton JR. Transmissible gastroenteritis: Loureiro ECB. Longitudinal study of ro- zel I, Duarte E. Rotavirus viral RNA elec- sodium transport in the intestinal epithe- tavirus infections among children from trophoresis in hospitalized infants with di- lium during the course of viral enteritis. Belém, Brazil. Epidemiol Infect 1989;102: arrhea in Santiago, Chile. Pediatr Res 1982; Gastroenterology 1977;72:457–461. 129–144. 16:29–30. 55. Kohler T, Erben U, Wiedersberg H, Bannert 69. Koopman JS, Monto AS The Tecumseh 83. Figueroa G, Araya M, Ibanez S, Clerc N, N. Histologische Befunde der Dunndarm- study. XV. Rotavirus infection and patho- Brunser O. Enteropathogens associated schleimhaut bei Rotavirusinfektionen im genicity. Am J Epidemiol 1989;130:750–759. with acute diarrhea in hospitalized infants. Sauglings-und Kleinkindalter. Kinderartl 70. Haffejee I. Neonatal rotavirus infections. J Pediatr Gastroenterol Nutr 1986;5: 226–231. Prax 1990;58:323–327. Rev Infect Dis 1991;13:957–962. 84. Hieber JP, Shelton S, Nelson JD, Leon J, 56. Lundgren O, Peregrin AT, Persson K, Kor- 71. Gusmão RHP, Mascarenhas JDP, Gabbay Mohs E. Comparison of human rotavirus dasti S, Uhnoo I, Svensson L. Role of the en- YB, Lins-Lainson Z, Ramos FLP, Monteiro disease in tropical and temperate settings. teric nervous system in the fluid and elec- TAF, et al. Rotaviruses as a cause of noso- Am J Dis Child 1978;132:853–858. trolyte secretion of rotavirus diarrhea. comial, infantile diarrhoea in Northern 85. Mata L, Simhon A, Urrutia JJ, Kronmal RA, Science 2000;287:491–495. Brazil. Mem Inst Oswaldo Cruz 1995;90: Fernandez R, Garcia B. Epidemiology of 57. Bartlett AV 3d, Bednarz-Prashad AJ, 743–749. rotaviruses in a cohort of 45 Guatamalan DuPont HL, Pickering LK Rotavirus gas- 72. Pacini DL, Brady MT, Budde CT, Connell Mayan Indian children observed from birth troenteritis. Ann Rev Med 1987;38:399–415. MJ, Hamparian VV, Hughes JH. Nosoco- to the age of three years. J Infect Dis 1983; 58. Bass DM, Greenberg HB. Group A ro- mial rotaviral diarrhea: pattern of spread 148:452–461. taviruses. In: Blaser MJ, Smith PD, Ravdin on wards in a children’s hospital. J Med 86. Suzuki H, Aman Y, Kinebuchi H, Gutierrez JI, Greenberg HB, Guerrant RL, eds. Infec- Virol 1987;23:359–366. VE, Davila A, Lopez J, et al. Rotavirus in- tions of the gastrointestinal tract. New 73. Rodriguez WJ, Kim HW, Brandt CD, Gard- fection in children with acute gastroen- York: Raven Press; 1995. pp. 967–982. ner MK, Parrott RH. Use of electrophoresis teritis in Ecuador. Am J Trop Med Hyg 59. Velazquez FR, Calva JJ, Guerrero ML, Mass of RNA from human rotavirus to establish 1981;30:293–294. D, Glass RI, Pickering LK, et al. Cohort the identity of strains involved in outbreaks 87. Pickering LK, Evans DJ Jr, Munoz O, study of rotavirus serotype patterns in in a tertiary care nursery. J Infect Dis 1983; DuPont HL, Coello-Ramirez P, Vollet JJ, et symptomatic and asymptomatic infections 148:34–40. al. Prospective study of enteropathogens in Mexican children. Ped Infect Dis J 1993; 74. Black RE, Brown KH, Becker S, Alim AR, in children with diarrhea in Houston and 12:54–61. Huq I. Longitudinal studies of infec- Mexico. J Pediatr 1978;93:383–388. 60. Isolauri E, Juntunen M, Wiren S, Vuorinen tious diseases and physical growth of chil- 88. Puerto FI, Polanco-Marín GG, Puerto-Solís P, Koivula T. Intestinal permeability dren in rural Bangladesh II. Incidence of MR, Ortega-Acosta G, Góngora-Biachi RA. changes in acute gastroenteritis: effects diarrhoea and association with known Diarrea infantil aguda por rotavirus en una of clinical factors and nutritional man- pathogens. Am J Epidemiol 1982;115: población pediátrica de Mérida, Yucatán, agement. J Ped Gastroenterol Nutr 1989;8: 315–324. México. Bol Med Hosp Infant Mex 1989;46: 466–473. 75. Pérez-Schael I. The impact of rotavirus 171–174. 61. Rodriguez WJ, Kim HW, Brandt CD, disease in Venezuela. J Infect Dis 1996; 174 89. Grinstein S, Gomez JA, Bercovich JA, Bis- Schwartz RH, Gardner MK, Jeffries B, et al. Suppl 1:S19–S21. cotti EL. Epidemiology of rotavirus infec- Fecal adenoviruses from a longitudinal 76. Rodriguez WJ, Kim HW, Brandt CD, tion and gastroenteritis in prospectively study of families in metropolitan Washing- Schwartz RH, Gardner MK, Jeffries B, et al. monitored Argentine families with young ton, D.C.: laboratory, clinical and epidemi- Longitudinal study of rotavirus infection children. Am J Epidemiol 1989;130:300–308. ologic observations. J Pediatr 1985;107: and gastroenteritis in families served by a 90. Araya M, Figueroa G, Espinoza J, Mon- 514–520. pediatric medical practice: clinical and epi- tesinos N, Spencer E, Brunser O. Acute di- 62. Santosham M, Burns B, Nadkarni V, Foster demiological observations. Pediatr Infect arrhoeal disease in children under 7 years S, Garrett S, Croll L, et al. Oral rehydration Dis J 1987;6:170–176. of age in a peri-urban slum of Santiago, therapy for acute diarrhea in ambulatory 77. Muchinik GR, Grinstein S, Plaza A. Ro- Chile. J Hyg 1985;95:57–67. children in the United States: a double- tavirus infection in children hospitalized 91. Beards GM, Desselberger U, Flewett TH. blind comparison of four different solu- for diarrhoea in Argentina. Ann Trop Pae- Temporal and geographical distribution of tions. Pediatrics 1985;76:159–166. diatr 1981;1:167–173. human rotavirus serotypes, 1983 to 1988. J 63. Riepenhoff-Talty M, Offor E, Klossner K, 78. Vergara M, Quiroga M, Grenon S, Villalba Clin Microbiol 1989;27:2827–2833. Kowalski E, Carmody PJ, Ogra PL. Effect of V, Pegels E, Chade M, et al. Identificación 92. Brussow H, Sidoti J, Barclay D, Sotek J, malnutrition on rotavirus infection in mice. de enteropatógenos en la diarrea de la Dirren H, Freire HB. Prevalence and Pediatr Res 1985;19:1250–1253. infancia y un estudio realizado en la ciudad serotype specificity of rotavirus antibodies 64. Biritwum RB, Isomura S, Yamaguchi H, de Posadas, Misiones, República Argentina. in different age groups of Ecuadorian in- Toba M, Mingle JA. Seroepidemiological Rev Latinoam Microbiol 1992;34: 71–75. fants J Infect Dis 1990;162:615–620. study of rotavirus infection in rural Ghana. 79. Coiro JR, De Almeida Neto AJ, Heuser MC, 93. Gomez J, Estes MK, Matson DO, Bellinzoni Ann Trop Paediatr 1984;4:237–240. Bendati MM, Vasconcellos VL Acute enteri- R, Alvarez A, Grinstein S. Serotyping of

326 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America human rotaviruses in Argentina by ELISA sequent rotavirus diarrhea. J Infect Dis rotavirus infections. J Infect Dis 1993;167: with monoclonal antibodies. Arch Virol 1993;168:282–287. 577–583. 1990;112:249–259. 108. Bernstein DI, Smith VE, Sander DS, Pax 121. Offit PA, Clark HF. Vaccines for enteric 94. Gonzalez-Losa MR, Puerto-Solis M, KA, Schiff GM, Ward RL. Evaluation of viral pathogens. In: Blaser MJ, Smith PD, Polanco-Marin GG, Peniche-Rodriguez R, WC3 rotavirus vaccine and correlates of Ravdin JI, Greenberg HB, Guerrant RL, eds. Puerto FI. Frecuencia de serotipos G de ro- protection in healthy infants. J Infect Dis Infections of the gastrointestinal tract. New tavirus aislados de niños con diarrea en 1990;162:1055–1062. York: Raven Press; 1995. pp. 1471–1478. Mérida, Yucatán, México. Rev Invest Clin 109. Chiba S, Yokoyama T, Nakata S, Morita Y, 122. Offit PA. Host factors associated with pro- 1994;46:215–219. Urasawa T, Taniguchi K, et al. Protective tection against rotavirus disease: the skies 95. Linhares AC, Gabbay YB, Mascarenhas effect of naturally acquired homotypic and are clearing. J Infect Dis 1996;174 Suppl 1: JDP, Freitas RB, Flewett TH, Beards GM. heterotypic rotavirus antibodies. Lancet S59–S64. Epidemiology of rotavirus subgroups and 1986;2(8504):417–421. 123. Offit PA, Dudzik KI. Rotavirus-specific cy- serotypes in Belém, Brazil: a three-year 110. De Champs C, Laveran H, Peigue-Lafeuille totoxic T lymphocytes cross-react with tar- study. Ann Virol (Inst. Pasteur) 1988;139: H, Chambon M, Demeocq F, Gaulme J, et get cells infected with different rotavirus 89–99. al. Sequential rotavirus infections: charac- serotypes. J Virol 1988;62:127–131. 96. Ward RL, Clemens JD, Sack DA, Knowlton terization of serotypes and electrophero- 124. Dhakarul T, Rott L, Greenberg HB. Recov- DR, McNeal MM, Huda N, et al. Culture types. Res Virol 1991;142:39–45. ery from chronic rotavirus infection in mice adaptation and characterization of group A 111. Reves RR, Mohassin MM, Midthun K, with severe combined immunodeficiency: rotaviruses causing diarrheal illnesses in Kapikian AZ, Naguib T, Zaqui AM, et al. virus clearance mediated by adoptive Bangladesh from 1985 to 1986. J Clin Mi- An observational study of naturally ac- transfer of immune CD8+ T lymphocytes. J crobiol 1991;29:1915–1923. quired immunity to rotaviral diarrhea in a Virol 1990;64:4375–4382. 97. White L, Garcia D, Boher Y, Blanco M, cohort of 363 Egyptian children. Calcula- 125. Franco MA, Feng N, Greenberg HB. Molecu- Pérez M, Romer H, et al. Temporal distri- tion of risk for second episodes using age- lar determinants of immunity and patho- bution of human rotavirus serotypes 1, 2, 3 specific person-years of observation. Am J genicity of rotavirus infection in the mouse and 4 in Venezuelan children with gas- Epidemiol 1989;130:981–988. model. J Infect Dis 1996;174 Suppl 1: S47–S50. troenteritis during 1979–1989. J Med Virol 112. Velázquez FR, Matson DO, Calva JJ, Guer- 126. Ramsey AJ, Ruby J, Ramshaw IA. A case 1991;34:79–84. rero L, Morrow AL, Cartel-Campbell S, et for cytokines as effector molecules in the 98. Gouvea V, de Castro L, Timenetsky MC, al. Rotavirus infection in infants as protec- resolution of virus infection. Immunol Greenberg H, Santos N. Rotavirus serotype tion against subsequent infections. N Engl J Today 1993;14:155–157. G5 associated with diarrhea in Brazilian chil- Med 1996;335:1022–1028. 127. Offit PA, Clark HF. Maternal antibody- dren. J Clin Microbiol 1994;32:1408–1409. 113. Arias CF, López S, Mascarenhas JDP, mediated protection against gastroenteritis 99. Leite JP, Alfieri AA, Woods PA, Glass RI, Romero P, Cano P, Gabbay YB, et al. Neu- due to rotavirus in newborn mice is depen- Gentsch JR. Rotavirus G and P types circu- tralizing antibody immune response in dent on both serotype and titer of antibody. lating in Brazil: characterization by RT- children with primary and secondary rota- J Infect Dis 1985;152:1152–1158. PCR, probe hybridization, and sequence virus infections. Clin Diagn Lab Immunol 128. Yolken RH, Losonsky GA, Vonderfecht S, analysis. Arch Virol 1996;141:2365–2374. 1994;1:89–94. Leister F, Wee SB. Antibody to human ro- 100. Timenetsky MC, Santos N, Gouvea V. Sur- 114. Gerna G, Sarasini A, Torsellini N, Torre D, tavirus in cow’s milk. N Engl J Med vey of rotavirus G and P types associated Parea M, Battaglia M. Group- and type-spe- 1985;312:605–610. with human gastroenteritis in São Paulo, cific serologic response in infants and chil- 129. Zheng BJ, Ma GZ, Tam JSL, Lo SKF, Hon Brazil, from 1986 to 1992. J Clin Microbiol dren with primary rotavirus infections and M, Lam BCC, et al. The effects of maternal 1994;32:2622–2624. gastroenteritis caused by a strain of known antibodies on neonatal rotavirus infection. 101. Bishop RF, Barnes GL, Cipriani E, Lund JS. serotype. J Infect Dis 1990;161:1105–1111. Pediatr Infect Dis J 1991;10:865–868. Clinical immunity after neonatal rotavirus 115. Green KY, Kapikian AZ. Identification of 130. Yolken RH, Peterson JA, Vonderfecht SL, infection: a prospective, longitudinal study VP7 epitopes associated with protection Fouts ET, Midthun K, Newburg DS. Hu- in young children. N Engl J Med 1983;309: against human rotavirus illness or shed- man milk mucin inhibits rotavirus replica- 72–76. ding in volunteers. J Virol 1992;66:548–553. tion and prevents experimental gastroen- 102. Ward Rl. Mechanisms of protection against 116. O’Ryan ML, Matson DO, Estes MK, Picker- teritis. J Clin Invest 1992;90:1984–1991. rotavirus in humans and mice. J Infect Dis ing LK. Anti-rotavirus G type-specific and 131. Clemens JD, Rao M, Ahmed F, Ward R, 1996;174 Suppl 1:S51–S58. isotype-specific antibodies in children with Huda S, Chakraborty J, et al. Breastfeeding 103. Hoshino Y, Saif LJ, Sereno MM, Chanock natural rotavirus infections. J Infect Dis and the risk of life-threatening rotavirus di- RM, Kapikian AZ. Infection immunity of 1994;169:504–511. arrhea: prevention or postponement? Pedi- piglets to either VP3 or VP7 outer capsid 117. Coulson BS, Grimwood K, Masendicz PJ, atrics 1993;92:680–685. protein confers resistance to challenge with Lund JS, Mermelstein N, Bishop RF, et al. 132. McLean BS, Holmes IH. Effects of antibod- a virulent rotavirus bearing the corre- Comparison of rotavirus immunoglobulin ies, trypsin and trypsin inhibitors on sus- sponding antigen. J Virol 1988;62:744–748. A coproconversion with other indices of ceptibility of neonates to rotavirus infec- 104. Ward RL, Kapikian AZ, Goldberg KM, rotavirus infection in a longitudinal study tion. J Clin Microbiol 1981;13:22–29. Knowlton DR, Watson MW, Rappaport R. in childhood. J Clin Microbiol 1990;28: 133. Schaller JP, Saif LJ, Cordle CT, Candler E Jr, Serum neutralizing-antibody titers com- 1367–1374. Winship TR, Smith KL. Prevention of hu- pared by plaque reduction and enzyme- 118. Clemens JD, Ward RL, Rao MR, Sack DA, man rotavirus induced diarrhea in gnotobi- linked immunosorbent assay-based neu- Knowlton DR, Van Loon FPL, et al. Sero- otic piglets using bovine antibody. J Infect tralization assays. J Clin Microbiol 1996;34: epidemiologic evaluation of antibodies to Dis 1992;165:623–630. 983–985. rotavirus as correlates of risk of clinically 134. Glass RI, Gentsch J, Smith JC. Rotavirus 105. McNabb PC, Tomasi TB. Host defense significant rotavirus diarrhea in rural vaccine: success by reassortment? Science mechanisms at mucosal surfaces. Annu Bangladesh. J Infect Dis 1992;165:161–165. 1994;265:1389–1390. Rev Microbiol 1981;35:477–496. 119. Coulson BS, Grimwood K, Hudson IL, 135. Snodgrass DR, Wells PW. Rotavirus infec- 106. Mestecky J, McGhee JR. Immunoglobulin A Barnes GL, Bishop RF. Role of coproanti- tion in lambs: studies on passive protection. (IgA): molecular and cellular interactions body in clinical protection of children dur- Arch Virol 1976;52:201–205. involved in IgA biosynthesis and immune ing reinfection with rotavirus. J Clin Micro- 136. Estes MK. Advances in molecular biology: response. Adv Immunol 1987;40:153–245. biol 1992;30:1678–1684. impact on rotavirus vaccine development. J 107. Bhan MK, Lew JF, Sazawal S, Das BK, 120. Matson DO, O’Ryan ML, Herrera I, Pick- Infect Dis 1996;174 Suppl 1:S37–S46. Gentsch JR, Glass RI. Protection conferred ering LK, Estes MK. Fecal antibody re- 137. Mebus CA, White RG, Bass EP, Twiehaus by neonatal rotavirus infection against sub- sponses to symptomatic and asymptomatic MJ. Immunity to neonatal calf diarrhea

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 327 virus. J Am Vet Med Assoc 1973; 163: reassortant vaccine for infants. J Infect Dis (rhesus rotavirus strain) in children. Pedi- 880–883. 1996;174 Suppl 1:S73–80. atrics 1987;80:473–480. 138. Wyatt RG, Yolken RH, Urrutia JJ, Mata L, 151. Garbarg-Chenon A, Fontaine J-L, Lasfar- 164. Bernstein DI, Glass RI, Rodgers G, David- Greenberg HB, Chanock RM, et al. Diar- gues G, Clark HF, Guyot J, Le Moing G, et son BL, Sack DA. Evaluation of rhesus rhea associated with rotavirus in rural al. Reactogenicity and immunogenicity of rotavirus monovalent and tetravalent re- Guatemala. A longitudinal study of 24 in- rotavirus WC3 vaccine in 5–12 month old assortant vaccines in US children. JAMA fants and young children. Am J Trop Med infants. Res Virol 1989;140:207–217. 1995;273:1191–1196. Hyg 1979;28:325–328. 152. Georges-Courbot MC, Monjes J, Siopathis 165. Christy C, Madore HP, Pichichero ME, 139. Clark HF, Borian FE, Bell LM, Modesto K, MR, Roungou JB, Gresenguet G, Bellec Gala C, Pancus P, Vosefski D, et al. Field Gouvea A, Plotkin AS. Protective effect of Bouquety C, et al. Evaluation of the efficacy trial of rhesus rotavirus vaccine in infants. WC3 vaccine against rotavirus diarrhea in of a low passage bovine rotavirus in chil- Pediatr Infect Dis J 1988;7:645–650. infants during a predominantly serotype 1 dren in Central Africa. Res Virol 1991;142: 166. Lanata CF, Midthun K, Black RE, Butron rotavirus season. J Infect Dis 1988;158: 405–411. B, Huapaya A, Penny ME, et al. Safety, 570–587. 153. Stuker G, Oshiro L, Schmidt NH. Antigenic immunogenicity, and protective efficacy 140. Hoshino Y, Sereno MM, Midthun K, Flores comparisons of two new rotaviruses from of one and three doses of the tetrava- J, Kapikian AZ, Chanock RM. Independent rhesus monkeys. J Clin Microbiol 1980;11: lent rhesus rotavirus vaccine in infants segregation of two antigenic specificities 202–203. in Lima, Peru. J Infect Dis 1996b; 174: (VP3 and VP7) involved in neutralization 154. Hoshino Y, Kapikian AZ, Chanock RM. Se- 268–275. of rotavirus infectivity. Proc Ntl Acad Sci lection of cold-adapted mutants of human 167. Madore HP, Christy C, Pichichero M, Long USA 1985;82:8701–8704. rotaviruses that exhibit various degrees of C, Pincus P, Vosefsky D, et al. Field trial of 141. Vesikari T, Isolauri E, Delem A, D’Hondt E, growth restriction in vitro. J Virol 1994;68: rhesus rotavirus or human-rhesus rota- André FE, Zissis G. Immunogenicity and 7598–7602. virus reassortant vaccine of VP7 serotype 3 safety of live oral attenuated bovine rota- 155. Nishikawa K, Fukuhara N, Liprandi N, or 1 specificity in infants. J Infect Dis 1992; virus vaccine strain RIT 4237 in adults Green K, Kapikian AZ, Chanock RM, et al. 166:235–243. and young children. Lancet 1983;2(8354): VP4 protein of porcine rotavirus strain 168. Gothefors L, Wadell G, Juto P, Taniguchi K, 807–811. OSU expressed by a baculovirus recombi- Kapikian AZ, Glass RI. Prolonged efficacy 142. Vesikari T, Isolauri E, D’Hondt E, Delem A, nant induces neutralizing antibodies. Virol- of rhesus rotavirus vaccine in Swedish chil- André FE, Zissis G. Protection of infants ogy 1989;173:631–637. dren. J Infect Dis 1989;159:753–757. against rotavirus diarrhoea by RIT 4237 at- 156. Wallace RE, Vasington PJ, Petricciani JC, 169. Pérez-Schael I, Garcia D, Gonzalez M, Gon- tenuated bovine rotavirus strain vaccine. Hopps HE, Lorenz DE, Kadanka Z. Devel- zalez R, Daoud N, Pérez M, et al. Prospec- Lancet 1984;1(8384):977–981. opment of a diploid cell line from fetal rhe- tive study of diarrheal diseases in Venezue- 143. Vesikari T, Ruuska T, Delem A, André FE. sus monkey lung for virus vaccine produc- lan children to evaluate the efficacy of Neonatal rotavirus vaccination with RIT tion. In Vitro 1973;8:323–332. rhesus rotavirus vaccine. J Med Virol 1990; 4237 bovine rotavirus vaccine: a prelimi- 157. Kapikian AZ, Hoshino Y, Chanock RM, 30:219–229. nary report. Pediatr Infect Dis J 1987; 6: Pérez-Schael I. Efficacy of quadrivalent rhe- 170. Ukae S, Nakata S, Adachi N, Kogawa K, 164–169. sus rotavirus-based human rotavirus vac- Chiba S. Efficacy of rhesus rotavirus vac- 144. Vesikari T, Isolauri E, Delem A, D’Hondt E, cine aimed at preventing severe rotavirus cine MMU-18006 against gastroenteritis André FE, Beards GM, et al. Clinical effi- diarrhea in infants and young children. J due to serotype 1 rotavirus. Vaccine 1994; cacy of the RIT 4237 live attenuated bovine Infect Dis 1996;174 Suppl 1:S65–S72. 12:933–939. rotavirus vaccine in infants vaccinated 158. Kapikian AZ, Vesikari T, Ruuska T, 171. Clark HF, Borian FE, Plotkin AS. Immune before a rotavirus epidemic. J Pediatr 1985; Madore HP, Christy C, Dolin R, et al. An protection of infants against rotavirus gas- 107:189–194. update on the “Jennerian” and modified troenteritis by a serotype 1 reassortant of 145. Hanlon P, Hanlon L, Marsh V, Byass P, “Jennerian” approach to vaccination of bovine rotavirus WC3. J Infect Dis 1990;161: Shenton F, Hassan-King M, et al. Trial of an infants and young children against rota- 1099–1104. attenuated bovine rotavirus vaccine (RIT virus diarrhea. Adv Exp Med Biol 1992;327: 172. Midthun K, Greenberg HB, Hoshino Y, 4237) in Gambian infants. Lancet 1987; 59–69. Kapikian AZ, Wyatt RG, Chanock RM. Re- 1(8546): 1342–1345. 159. Pérez-Schael I, González M, Daoud N, assortant rotaviruses as potential live ro- 146. De Mol P, Zissis G, Butzler J-P, Mut- Pérez M, Soto I, Garcia D, et al. Reacto- tavirus candidate vaccines. J Virol 1985;53: wewingabo A, Andre FE. Failure of live, genicity and antigenicity of the rhesus ro- 949–954. attenuated oral rotavirus vaccine. Lancet tavirus vaccine in Venezuelan children. J 173. Midthun K, Kapikian AZ. Rotavirus vac- 1986; 2(8498):108. Infect Dis 1987;155:334–338. cines: an overview. Clin Microbiol Rev 147. Santosham M, Letson GW, Wolff M, Reid 160. Lanata CF, Black RE, Flores J, Lazo F, 1996;9:423–434. R, Gahagan S, Adams R, et al. A field study Butron B, Linares A, et al. Immunogenicity, 174. Treanor JJ, Clark HF, Pichichero M, Christy of the safety and efficacy of two candidate safety and protective efficacy of one dose of C, Gouvea V, Shrager D, et al. Evaluation rotavirus vaccines in a Native American the rhesus rotavirus vaccine and serotype 1 of the protective efficacy of a serotype 1 population. J Infect Dis 1991;163:483–487. and 2 human-rhesus rotavirus reassortants bovine-human rotavirus reassortant vac- 148. Lanata CF, Black RE, del Aguila R, Gil A, in children from Lima, Peru. Vaccine 1996; cine in infants. Pediatr Infect Dis J 1995;14: Verastegui H, Gerna G, et al. Protection of 14:237–243. 301–397. Peruvian children against rotavirus diar- 161. Losonsky GA, Rennels MB, Kapikian AZ, 175. Sereno MM, Gorziglia MI. The outer capsid rhea of specific serotypes by one, two, or Midthun K, Ferra PJ, Fortier DN, et al. protein VP4 of murine rotavirus Eb repre- three doses of the RIT 4237 attenuated Safety, infectivity, transmissibility and im- sents a tentative new P type. Virology 1994; bovine rotavirus vaccine. J Infect Dis 1989; munogenicity of rhesus rotavirus vaccine 199:500–504. 159:452–459. (MMU 18006) in infants. Ped Infect Dis 176. Clark HF, White CJ, Offit PA, Stinson D, 149. Clark HF, Furukawa T, Bell LM, Offit PA, 1986;5:25–29. Eiden J, Weaver S, et al. Preliminary evalu- Perrella PA, Plotkin AS. Immune response 162. Vesikari T, Ruuska T, Delem A, André FE. ation of safety and efficacy of quadrivalent of infants and children to low-passage Oral rotavirus vaccination in breast-and human-bovine reassortant rotavirus vac- bovine rotavirus (strain WC3). Amer J Dis bottle-fed infants aged 6 to 12 months. Acta cine Ped Res 1995;37:127A. Children 1986;140:350–356. Paediatr Scand 1986;75:573–578. 177. Clements-Mann ML, Makhene MK, 150. Clark HF, Offit PA, Ellis RW, Eiden JJ, Krah 163. Wright PF, Tajima T, Thompson J, Mrukowicz J, Wright PF, Hoshino Y, D, Shaw AR, et al. The development of Kokubun K, Kapikian A, Edwards K, et al. Midthun K, et al. Safety and immunogenic- multivalent bovine rotavirus (strain WC3) Evaluation of a candidate rotavirus vaccine ity of live attenuated human-bovine (UK)

328 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America reassortant rotavirus vaccines with VP7- 189. Joensuu J, Koskenniemi E, Pang X-L, Safety and immunogenicity of human rota- specificity for serotypes 1, 2, 3 or 4 in Vesikari T. Randomised placebo-controlled virus vaccine strain M37 in adults, children adults, children and infants. Vaccine 1999; trial of rhesus-human reassortant rotavirus and infants. J Infect Dis 1991;164:792–796. 17: 2715–2725. vaccine for prevention of severe rotavirus 204. Vesikari T, Ruuska T, Delem A, André FE, 178. Midthun K, Hoshino Y, Kapikian AZ, gastroenteritis. Lancet 1997;350:1205–1209. Beards GM, Flewett TH. Efficacy of two Chanock RM. Single gene substitution ro- 190. Vesikari T, Joensuu J. Review of rotavirus doses of RIT 4237 bovine rotavirus vaccine tavirus reassortants containing the major vaccine trials in Finland. J Infect Dis 1996; for prevention of rotavirus diarrhoea. Acta neutralization protein (VP7) of human ro- 174 Suppl 1:S81–S87. Paediatr Scand 1991;80:173–180. tavirus serotype 4. J Clin Microbiol 1986;24: 191. Linhares AC, Gabbay YB, Mascarenhas 205. Bishop RF, Hewstone AS, Davidson GP, 822–826. JDP, de Freitas RB, Oliveira CS, Bellesi N, Townley RR, Holmes IH, Ruck BJ. An epi- 179. Vesikari T, Ruuska T, Green K, Flores J, et al. Immunogenicity, safety and efficacy demic of diarrhea in human neonates Kapikian AZ. Protective efficacy against of rhesus-human, reassortant rotavirus vac- involving a reoviruslike agent and “entero- serotype 1 rotavirus diarrhea by live oral cine in Belém, Brazil. Bull WHO 1996;74: pathogenic” serotypes of Escherichia coli. J rhesus-human reassortant rotavirus vac- 491–500. Clin Pathol 1976;29:46. cines with human rotavirus VP7 serotype 1 192. Linhares AC, Lanata CF, Hausdorff WP, 206. Bernstein DI, Sack DA, Rothstein E, or 2 specificity. Pediatr Infect Dis J 1992;11: Gabbay YB, Black RE. Reappraisal of the Reisinger K, Smith VE, O’Sullivan D, et al. 535–542. Peruvian and Brazilian lower titer tetrava- Efficacy of live, attenuated, human rota- 180. Rennels MB, Wasserman SS, Glass RI, lent rhesus-human reassortant rotavirus virus vaccine 89–12 in infants: a randomised Keane V. Comparison of immunogenicity vaccine efficacy trials: analysis by severity placebo-controlled trial. Lancet 1999;354: and efficacy of rhesus rotavirus reassortant of diarrhea. Pediatr Infect Dis J 1999;18: 287–290. vaccines in breastfed and nonbreastfed 1001–1006. 207. Matsuno S, Murakami S, Takagi M, children. Pediatrics 1995;96:1132–1136. 193. Pérez-Schael I, Guntiñas MJ, Pérez M, Hayashi M, Inouye S, Hasegawa A, et al. 181. Ward RL, Bernstein DI. Lack of correlation Pagone V, Rojas AM, González R, et al. Cold-adaptation of human rotavirus. Virus between serum rotavirus antibody titers Efficacy of the rhesus rotavirus-based Res 1987;7:273–280. and protection following vaccination with quadrivalent vaccine in young children 208. Ijaz MK, Sabara MJ, Frenchik PJ, Babiuk reassortant RRV vaccines. US Rotavirus in Venezuela. N Engl J Med 1997; 337: LA. Assessment of intestinal damage in ro- Vaccine Efficacy Group. Vaccine 1995;13: 1181–1187. tavirus infected neonatal mice by a D-xylose 1226–1232. 194. United States, Centers for Disease Control absorption test. J Virol Methods 1987;18: 182. Dagan R, Kassis I, Sarov B, Midthun K, and Prevention. Intussusception among 153–157. Davidson B, Vesikari T, et al. Safety and im- recipients of rotavirus vaccine—United 209. Sheridan JF, Smith CC, Manak MM, Aure- munogenicity of oral tetravalent human- States, 1998–1999. MMWR 1999;48:577–581 lian L. Prevention of rotavirus induced rhesus reassortant rotavirus vaccine in neo- 195. Committee on Infectious Diseases, 1999– diarrhea in neonatal mice born to dams im- nates. Pediatr Infect Dis 1992;11: 991–996. 2000. Possible association of intussuscep- munized with empty capsids of simian ro- 183. Flores J, Pérez-Schael I, Blanco M, White L, tion with rotavirus vaccination. Pediatrics tavirus SA11. J Infect Dis 1984;149:434–438. Garcia D, Vilar M, et al. Comparison of re- 1999;104:575. 210. Andrew ME, Boyle DB, Coupar BEH, Whit- actogenicity of M37 rotavirus vaccine and 196. World Health Organization. Report of the field PL, Both GW, Bellamy AR. Vaccinia rhesus-rotavirus-based quadrivalent vac- Meeting on Future Directions for Rotavirus virus recombinants expressing the SA11 cine. Lancet 1990;336:330–334. Vaccine Research in Developing Countries. rotavirus VP7 glycoprotein gene induce 184. Ing DJ, Glass RI, Woods PA, Simonetti M, Geneva, 9–11 February, 2000. Geneva: WHO; serotype-specific neutralising antibodies. J Pallansch MA, Wilcox WD, et al. Immuno- 2000. (Publication WHO/V&B/00.23). Virol 1987;61:1054–1060. genicity of tetravalent rhesus rotavirus vac- 197. United States, Centers for Disease Control 211. Reeves PR, Johnson MA, Holmes IH, Dyall- cine administered with buffer and oral po- and Prevention. Withdrawal of rotavirus Smith ML. Expression of rotavirus VP7 liovirus vaccine. Am J Dis Child 1991;145: vaccine recommendation. MMWR 1999; antigens in fusions with bacterial proteins. 892–897. 282:1007. Res Microbiol 1990;141:1019–1025. 185. Pérez-Schael I, Blanco M, Garcia D, White 198. Rennels MB, Parashar UD, Holman RC, Le 212. Conner ME, Matson DO, Estes MK. Rota- L, Alfonzo E, Crespo I, et al. Evaluation of CT, Chang H-G, Glass RI, et al. Lack of ap- virus vaccines and vaccination potential. In: the antigenicity and reactogenicity of vary- parent association between intussusception Ramig RF, ed. Rotaviruses. Berlin: Springer- ing formulations of the rhesus rotavirus- and wild or vaccine rotavirus infection. Ped Verlag; 1994. pp. 286–337. based quadrivalent and the M37 rotavirus Infect Dis J 1998;17:924–925. 213. Andrew ME, Boyle DB, Coupar BEH, vaccine candidates. J Med Virol 1994;42: 199. Foster RH, Wagstaff AJ. Tetravalent human- Reddy D, Bellamy AR, Both GW Vaccinia- 330–337. rhesus reassortant rotavirus vaccine: a re- rotavirus VP7 recombinants protect mice 186. Rennels MB, Glass RI, Dennehy PH, Bern- view of its immunogenicity, tolerability and against rotavirus-induced diarrhoea. Vaccine stein DI, Pichichero ME, Zito ET, et al. protective efficacy against paediatric rota- 1992;10:189–191. Safety and efficacy of high-dose rhesus- virus gastroenteritis. Biodrugs 1998;9: 214. Andrew ME, Boyle DB, Whitfield PL, Lock- human reassortant rotavirus vaccines— 155–178. ett LJ, Anthony ID, Bellamy AR, et al. The report of the National Multicenter Trial. 200. Konno T, Suzuki H, Kutsuzawa T, Imai A, immunogenicity of VP7, a rotavirus anti- United States Rotavirus Vaccine Efficacy Katsushima M, Sakamoto M, et al. Human gen resident in the endoplasmic reticulum, Group. Pediatrics 1996;97:7–13. rotavirus infection in infants and young is enhanced by cell surface expression. J 187. Santosham M, Moulton LH, Reid R, Croll J, children with intussusception. J Med Virol Virol 1990;64:4776–4783. Weatherholt R, Ward R, et al. Efficacy and 1978; 2: 265–269. 215. Conner ME, Zarley CD, Hu B, Parsons S, safety of high-dose rhesus-human reassor- 201. Mulcahy DL, Kamath KR, de Silva LM, Drabinsky D, Greiner S, et al. Virus-like tant rotavirus vaccine in Native American Hodges S, Carter IW, Cloonan MJ. A two- particles as rotavirus subunit vaccine. J In- populations. J Pediatr 1997;131:632–638. part study of the aetiological role of rota- fect Dis 1996;174 Suppl 1:S88–S92. 188. Pichichero ME, Marsocci SM, Francis AB, virus in intussusception. J Med Virol 1982; 216. Crawford SE, Labbé M, Cohen J, Burroughs Green JL, Disney FA, Rennels MB, et al. 9:51–55. MH, Zhou Y, Estes MK. Characterization of A comparative evaluation of the safety 202. Nicolas JC, Ingrand D, Fortier B, Bricout F. virus-like particles produced by the expres- and immunogenicity of a single dose of un- A one-year virological survey of acute sion of rotavirus capsid proteins in insect buffered oral rhesus/human reassortant intussusception in childhood. J Med Virol cells. J Virol 1994;68:5945–5952. serotypes 1, 2 and 4 and combined (tetrava- 1982;9:267–271. 217. Estes MK, Crawford SE, Penaranda EM, lent) vaccines in healthy infants. Vaccine 203. Midthun K, Halsey NA, Jett-Goheen M, Petrie BL, Burns JW, Chan W-K, et al. 1993;11:747–753. Clements ML, Steinhoff M, King JC, et al. Synthesis and immunogenicity of the ro-

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 329 tavirus major capsid antigen using a bac- genicity and efficacy of rotavirus vaccines. J vaccine (OPV) in Thai infants. Vaccine 1995; ulovirus expression system. J Virol 1987; 61: Infect Dis 1996;174 Suppl 1:S107–S111. 13:168–174. 1488–1494. 229. Glass RI, Ing DJ, Stoll BJ, Ing RT. Immune 239. World Health Organization Collaborative 218. Bloom BR. Vaccines for the Third World. response to rotavirus vaccine among breast- Study Group on Oral Poliovirus Vaccine. Nature 1989;342:115–120. fed and nonbreast-fed children. In: Mestecky Factors affecting the immunogenicity of 219. Both GW, Lockett LJ, Janardhana V, Ed- J, ed. Immunology of milk and the neonate. oral poliovirus vaccine: a prospective eval- wards SJ, Bellamy AR, Graham FL, et al. New York: Plenum Press; 1991. p. 249–253. uation in Brazil and Gambia. J Infect Dis Protective immunity to rotavirus-induced 230. Pichichero ME. Effect of breast-feeding on 1995;171:1097–1106. diarrhoea is passively transferred to new- oral rhesus rotavirus vaccine seroconver- 240. World Health Organization, United Nations born mice from naive dams vaccinated sion: a metaanalysis. J Infect Dis 1990;162: Children’s Fund. State of the world’s vac- with a single dose of a recombinant adeno- 753–755. cines and immunization. Geneva, Switzer- virus expressing rotavirus VP7sc. Virology 231. Ceyhan M, Kanra G, Seçmeer G, Midthun land: WHO; 1996. (Publication WHO/GPV/ 1993;193:940–950. K, Davidson BL, Zito ET, et al. Take of 96.04). 220. Gorziglia M, Kapikian AZ. Expression of rhesus-human reassortant rotavirus vac- 241. Clemens J, Brenner R, Rao M, Tafari N, the OSU rotavirus outer capsid protein VP4 cine in breast-fed infants. Acta Paediatr Lowe C. Evaluating new vaccines for de- by adenovirus recombinant. J Virol 1992;66: 1993;82:223–227. veloping countries. Efficacy or effective- 4407–4412. 232. Vodopija I, Baklaic Z, Vlatkovic R, Bogaerts ness. JAMA 1996;275:390–397. 221. Gunn PR, Sato F, Powell KFH, Bellamy AR, H, Delem A, André FE. Combined vac- 242. Griffiths RI, Anderson GF, Powe NR, Oliv- Napier JR, Harding DRK, et al. Rotavirus cination with live oral polio vaccine and the eras E, Herberts RJ, Grant CC, et al. Eco- neutralizing protein VP7: antigenic deter- bovine rotavirus RIT 4237 strain. Vaccine nomic impact of immunization against minants investigated by sequence analysis 1996;4:233–236. rotavirus gastroenteritis. Arch Pediatr Ado- and peptide synthesis. J Virol 1985;54: 233. Giammanco G, DeGrandi V, Lupo L, Mis- lesc Med 1995;149:407–414. 791–797. tretta A, Pignato S, Teuween D, et al. Inter- 243. Smith JC, Haddix AC, Teutsch SM, Glass 222. Arias CF, Garcia G, López S. Priming for ference of oral poliovirus vaccine on RIT RI. Cost-effectiveness analysis of a rota- rotavirus neutralizing antibodies by a VP4 4237 oral rotavirus vaccine. Eur J Epidemiol virus immunization program for the protein-derived synthetic peptide. J Virol 1988;4:233–236. United States. Pediatrics 1995;96:609–615. 1989;63:5393–5398. 234. Ho MS, Floyd RL, Glass RI, Pallansch MA, 244. Tucker AW, Haddix AC, Bresee JS, Holman 223. Borras-Cuesta F, Petit-Camurdan A, Fedon Jones B, Hamby B, et al. Simultaneous ad- RC, Parashar UD, Glass RI. Cost-effective- Y. Engineering of immunogenic peptides ministration of rhesus rotavirus vaccine ness analysis of a rotavirus immunization by co-linear synthesis of determinants rec- and oral poliovirus vaccine: immunogenic- program for the United States. JAMA 1998; ognized by B and T cells. Eur J Immunol ity and reactogenicity. Pediatr Infect Dis J 259:1371–1376. 1987;17:1213–1215. 1989;8:692–696. 245. Bresee JS, Glass RI, Ivanoff B, Gentsch JR. 224. Herrmann JE, Chen SC, Fynan EF, Santoro 235. Jalil F, Zaman S, Carlsson B, Glass RI, Current status and future priorities for ro- JC, Greenberg HB, Wang S, et al. Protection Kapikian AZ, Mellander L, et al. Immuno- tavirus vaccine development, evaluation against rotavirus infections by DNA vac- genicity and reactogenicity of rhesus ro- and implementation in developing coun- cination. J Infect Dis 1996;174 (Suppl 1): tavirus vaccine given in combination with tries. Vaccine 1999;17:2207–2222. S93–S97. oral or inactivated poliovirus vaccine and 246. Orenstein WA, Hadler S, Kuritsky JN, 225. Webster RG, Fynan EF, Santoro JC, Robin- diphtheria-tetanus-pertussis vaccine. Trans Bernier RH. Rotavirus vaccines–from licen- son H. Protection of ferrets against influenza R Soc Trop Med Hyg 1991;85:292–296. sure to disease reduction. J Infect Dis 1996; challenge with a DNA vaccine to haemag- 236. Rennels MB, Ward RL, Mack ME, Zito ET. 174 (Suppl 1): S118–S124. glutinin. Vaccine 1994;12:1495–1498. Concurrent oral poliovirus and rhesus- 247. Henchal LS, Midthun K, Goldenthal KL. 226. World Health Organization, Centers for human reassortant rotavirus vaccination ef- Selected regulatory and scientific topics for Disease Control and Prevention, Children’s fects on immune responses to both vaccines candidate rotavirus vaccine development. J Vaccine Initiative. Report on Consensus and on efficacy of rotavirus vaccines. The Infect Dis 1996;174 (Suppl 1):S112–S117. Workshop: Rotavirus Vaccines for the Im- US Rotavirus Vaccine Efficacy Group. J In- 248. United States, Department of Health and munization of Children in Developing fect Dis 1996;173:306–313. Human Services. Points to consider in the Countries. Geneva, 9–10 January, 1997. 237. Rennels MB. Influence of breast-feeding manufacture and testing of monoclonal Geneva, Switzerland: WHO; 1997. and oral poliovirus vaccine on the immuno- antibody products for human use. Fed Reg 227. Pickering LK, Morrow AL, Herrera I, genicity and efficacy of rotavirus vaccines. J 1994;59:39571–39572. O’Ryan M, Estes MK, Guilliams SE, et al. Infect Dis 1996; 174(Suppl 1): S107–S111. Effect of maternal rotavirus immunization 238. Migasena S, Simasathien S, Samakoses R, on milk and serum antibody titers. J Infect Pitisuttiham P, Sangaroon P, van Steenis G, Dis 1995;172:723–728. et al. Simultaneous administration of oral Original manuscript received 12 March 1999 and ac- 228. Rennels MB. Influence of breast-feeding rhesus-human reassortant tetravalent (RRV- cepted for publication 21 August 1999; revised accepted and oral poliovirus vaccine on the immuno- TV) rotavirus vaccine and oral poliovirus version received 22 September 2000.

330 Linhares and Bresee • Rotavirus vaccines and vaccination in Latin America RESUMEN En el mundo, los rotavirus son responsables de más 125 millones de casos de gas- troenteritis infantil y de cerca de 1 millón de muertes al año, especialmente en los países en desarrollo. En comparación con otras medidas de control, la vacunación Vacunas y vacunación contra cuenta con mayores probabilidades de tener un impacto significativo en la incidencia rotavirus en América Latina de la enfermedad por rotavirus. La máxima incidencia de la diarrea por rotavirus ocurre entre los 6 y los 24 meses de edad. Sin embargo, en los países en desarrollo no son raros los casos en menores de 6 meses. Los serotipos G1 a G4 son responsables de la mayoría de los casos, pero hay datos que indican que en Brasil el tipo G5 está adquiriendo mayor importancia epidemiológica. Durante la infección natural por ro- tavirus se inducen respuestas inmunitarias tanto homotípicas como heterotípicas y la respuesta inmunitaria en la superficie de la mucosa intestinal constituye probable- mente el factor que mejor predice la inmunidad clínica. Con el objetivo principal de proteger a los niños frente a la diarrea con deshidra- tación potencialmente mortal, se han probado varios abordajes en el desarrollo de va- cunas contra los rotavirus. En los Estados Unidos de América (EUA) se aprobó y co- mercializó una vacuna antirrotavirus tetravalente recombinante de virus humanos y de Macaca mulatta (RRV-TV), pero posteriormente se retiró del mercado. En varios estudios se ha observado que la eficacia de la vacuna RRV-TV es mayor en los países desarrollados que en los países en desarrollo. En dos países de América Latina (Brasil y Perú) se han realizado estudios de campo con preparaciones de RRV- TV que contenían 4 104 unidades formadoras de placa (UFP) y se registraron tasas de eficacia protectora frente a todas las diarreas por rotavirus que oscilaron entre 18 y 35%. En otro estudio realizado en Venezuela con mayores dosis de RRV-TV (4 105 UFP/dosis), se obtuvo una tasa de eficacia de 48% frente a todas las diarreas por ro- tavirus y de 88% frente a las diarreas graves. Parece ser que la lactancia materna no interfiere con la eficacia de la RRV-TV si se administran tres dosis de la vacuna. Del mismo modo, la posible interferencia de la vacuna oral contra la poliomielitis con la vacuna contra los rotavirus puede ser contrarrestada si se administran tres dosis de esta última o se utilizan formulaciones con títulos elevados. Sin embargo, los en- terovirus silvestres pueden conducir al fracaso de la vacunación primaria contra los rotavirus en los países en desarrollo. Estudios realizados en Perú con la vacuna RRV- TV han mostrado una tendencia a la obtención de mayores tasas de eficacia de la va- cuna frente a los epidosios diarreicos causados únicamente por rotavirus (“puros”). Los análisis económicos realizados en los EE.UU. indican que una vacuna que cuesta menos de 9 dólares estadounidenses por dosis podría proporcionar ahorros netos en los costes médicos. Sin embargo, hasta la fecha no se han realizado estudios de coste-beneficio en los países en desarrollo. En el futuro, es posible que algunos países de América Latina adapten sus instalaciones de producción de vacunas an- tipoliomielíticas a la preparación de vacunas contra los rotavirus para uso humano. Un año después de que la RRV-TV fuera aprobada en los EE.UU. para la vacu- nación de lactantes, la ocurrencia de casos de intususcepción como reacción adversa a la vacuna condujo a su retirada del mercado. En este artículo se analizan las impli- caciones de esta medida, en especial en América Latina, entre ellas la necesidad de in- vestigar vacunas alternativas contra los rotavirus, particulamente mediante la rea- lización de ensayos clínicos paralelos en los países desarrollados y en desarrollo.

Rev Panam Salud Publica/Pan Am J Public Health 8(5), 2000 331