J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from
MEDICAL PRACTICE Adenocarcinoma of the Uterine Cervix
A nth()r~v F .Ierant, MJ)
Background: Adenocarcinoma of the uterine cervix is an increasingly common cervical neoplasm that has received little attention in the primary care literature. The purpose of this paper is to describe an illustrative case that provides an excellent opportunity to review the symptoms, diagnostic pitfalls, treatment options, and prognosis of this important disease. Methods: Case report is described, along with results of a literature review using MEDLINE and pertinent references from retrieved articles. Results: The relative incidence of cervical adenocarcinoma has risen from 5 to 10 percent of all cervical neoplasms in the 1950s to 10 to 20 percent in recent series. Some studies have also reported an increasing absolute incidence linked to widespread oral contraceptive use. The diethylstilbestrol-associated clear-cell variant accounts for only 2 to 3 percent of cases. About 10 percent of patients have only a nonbloody vaginal discharge. Cervical adenocarcinoma might be more easily missed on a Papanicolaou smear than squamous cell dysplasia and cancer, and it has no characteristic colposcopic appearance. The prognosis is excellent with early detection. Conclusions: Family physicians should maintain a high index of suspicion for cervical adenocarcinoma when symptoms suggest this disease regardless of Papanicolaou smear results (J Am Board Fam Pract 1997;10:36-42).
Adenocarcinoma of the uterine cervix, an uncom "adenocarcinoma" were combined using the mon neoplasm in the past, has increased in rela "AND" function. The ensuing set of references tive and perhaps absolute incidence in the last 20 was limited by the "human" and "English lan years. I-3D Although this trend is well-documented guage only" functions. Relevant articles were in gynecologic and oncology journals, it has not then selected by reviewing the titles and abstracts. been addressed in the family medicine literature. References from reviewed articles were also http://www.jabfm.org/ The prognosis for patients with cervical adeno checked for relevant citations not found on carcinoma is excellent when the disease is de MEDLINE. tected early, but diagnosis is difficult unless the primary care physician is vigilant when symptoms Case Report suggest this disease. The purpose of this review is A 34-year-old woman complained of a I-month to alert family physicians to the clinical signs, history of copious, clear vaginal discharge. Her symptoms, and diagnostic pitfalls of this increas menstrual cycles were regular, and there was no on 24 September 2021 by guest. Protected copyright. ingly important malignancy. history of intermenstrual bleeding or spotting. She noted no foul odor to the discharge and no Methods dysuria, urinary frequency, or urgency. There After a case report of cervical adenocarcinoma is were no risk factors for sexually transmitted dis described, the results of a computerized MED eases. She was not using any birth control, but she LINE search from 1966 to 1995 are discussed. did report using oral contraceptives continuously The MeSH headings "cervix neoplasms" and between the ages of 19 and 24 years. She did not smoke and had not been exposed to diethylstil bestrol in utero. Specimens for previous Papani Submitted, revised, 21 May 1996. colaou smears were adequate, and the results From the Department of Family and Community Medicine, were normal; the most recent one had been per Eisenhower Army lvleclical Center, Fort Gordon, Ga. Address formed 6 months earlier. reprint requests to Anthony E Jcrant, MD, Department of 1'.111\ ily and Com III unity Medicine, Eisenhower Army Medical Cen When examined vaginally, she had copious ter, Fort Cordon, GA 3(90). clear mucoid discharge at the introitus, which was
36 JABFP Jan.-Feb. 1<)97 Vol. 10 No.1 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from apparent even before speculum insertion. Find Six months after surgery, the patient continued ings on an external genital examination were unre to do well on hormone replacement therapy and markable. Examination using a speculum showed had no cancer recurrence at frequent follow-up a normal-appearing vaginal canal and a cervix that visits. appeared somewhat diffusely friable. A slightly raised, condyloma-like lesion was noted at about Adenocarcinoma 11 o'clock near the tranformation zone. Findings History on bimanual and rectovaginal examinations were According to data from case series before the unremarkable. Potassium hydroxide and saline 1970s, adenocarcinoma made up only 5 to 10 per wet preparation slides of the discharge showed no cent of all cervical malignancies. 3,8,1 1,16-18,21,24,27,28 abnormalities. Because there was no clear explana Thus, in primary care textbooks less attention has tion for the discharge after initial examination, the been focused on cervical adenocarcinoma than on patient was referred for colposcopy. The pre squamous cell carcinoma. 31 Since the mid-1970s, sumptive diagnosis was cervical condyloma and however, numerous case series from Scandinavia, human papillomavirus (HPV) infection. Europe, the United Kingdom, and the United During the colposcopic examination of the States have documented an increase in relative cervix, there was whitening of the raised lesion incidence, with figures ranging from 10 to 20 per with application of acetic acid solution. The col cent. 1-30 Some have attributed this increase to the poscopist remarked that the lesion, which was well-documented decrease in invasive squamous near the edge of the transformation zone, "looked cell cancer after the advent of mass screening with just like a wart." No abnormal vascular pattern Papanicolaou smears in the 1950s.9,32 Although was noted. Samples of the lesion and two other this association might account for an increasing mildly whitened areas were taken for biopsy, en relative incidence of cervical adenocarcinoma, a docervical curettage was performed, and a Papan number of reports also suggest an increasing abso icolaou smear was repeated. lute incidence for adenocarcinoma. 1,2,6,7,9,2 1,23,24,29 The Papanicolaou smear showed atypical glan Referral bias must be considered in series origi dular cells of undetermined significance but sug nating from gynecologic oncology centers but gesting a premalignant process. Findings from cannot account for similar findings in population the endocervical curettage showed normal cells. based studies.2,6,21,23 Pathologic examination of the wart-like lesion re http://www.jabfm.org/ vealed well-differentiated adenocarcinoma, ap Histopathology parently of cervical glandular origin. The depth The histologic classification of cervical adenocar of invasion of the neoplasm could not be deter cinomas is controversial, and several schemes ex mined because the biopsy specimen was too ist. Table 1 outlines a simple categorization that is superficial. The other biopsies were negative for useful for primary care clinicians.33 Up to 40 per any abnormality. cent of adenocarcinomas contain a mixture of The patient was referred immediately to a gy subtypes. Endocervical adenocarcinoma often oc on 24 September 2021 by guest. Protected copyright. necologic oncologist for further examination. M curs in association with squamous neoplasia, and ter a complete survey for metastases and review of tumors with invasive elements of both cell types the lesion, the oncologist diagnosed stage IE en docervical adenocarcinoma. After receiving coun seling and education about the risks and benefits Table 1. Simplified Classification of radiation and surgical therapies, the patient of Cervical Carcinoma. chose the latter treatment. She underwent a radi 1. Squamous cell carcinomas cal hysterectomy, bilateral salpingo-oophorec II. Adenocarcinomas tomy, and pelvic and para-aortic lymph node dis A. Endocervical section. Her surgical and postoperative course B. Endometroid was uneventful. Final review of her surgical C. Clear cell D. Adenoid cystic pathology specimens showed a small amount of III. Undifferentiated carcinomas residual cervical disease and absence of tumor at the surgical margins and in the lymph nodes. Adapted from Kilgore and Helm.!3
Adenocarcinoma of Cervix 37 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from are categorized as adenosquamous.H The most oral contraceptive pills, because the pill was first plausible explanation for the coexistence of these widely prescribed in the early 1960s. In these and cancers is that they arise from a suhcolumnar subsequent observational studies, an increased in stem or reserve cell that is capable of differentia cidence of cervical adenocarcinoma was first de tion to both squamolls ,mel glandular epithelium, tected approximately 15 years after the peak ages a theory supported by cell ultrastructural studies for oral contraceptive initiation, but in the ab of adenocarcinoma in situ (AIS).H Because most sence of better case-control studies or any cohort cases of invasive adenocarcinoma are diagnosed studies, the association remains speculative. In without a preceding diagnosis of AIS, in sharp terestingly, the patient described in the present contrast to the widely acknowledged progression case report was 34 years old, and she had used from dysplasia to carcinoma-in-situ to invasive oral contraceptives for 5 years starting at 19 years, cancer documented for squamous cell neoplasms, approximately 15 years before her cervical adeno there is some question whether AIS represents a carcinoma was diagnosed. precursor lesion to invasive adenocarcinoma. 35 It is relatively well-established that infection AIS is difficult to diagnose, however, because un with HPV incurs an increased risk of cervical ade like squamous cell carcinoma in situ, the patholo nocarcinoma, similar to the association between gist cannot use the basement membrane as a HPV and squamous cell neoplasms.34,35,37,38 readily identifiable boundary for measurement of Nearly all papers report HPV-18 as the predomi invasion. Thus, the unclear relationship between nant subtype found in adenocarcinoma as well as AIS and invasive adenocarcinoma could ref1ect adenocarci noma in si tu. HPV-16 has been iso the difficulty in diagnosing AIS rather than a lated less frequently. truly independent histologic development. Descriptive data on other risk factors for cervi The clear-cell variant includes neoplasms re cal adenocarcinoma are quite sparse. Asian race, lated to intrauterine exposure to diethylstil obesity, early age at first intercourse, and multiple hestrol. In utero diethylstilbestrol exposure, al sexual partners might confer an increased risk of though widely recognized as a risk factor by adenocarcinoma. 5,H,2o,33-35 Nulliparity and ciga family physicians, accounts for only 2 to 3 percent rette smoking have not been shown to increase of cervical adenocarcinomas.H The endometroid risk for this type of cervical cancer. The studies to subtype can be difficult to differentiate from en date are far from conclusive and often contradic dometrial carcinoma that has spread to the cervix tory, however. and can be diagnosed only if the endometrium is http://www.jabfm.org/ normal. 3+ Diagnosis Cervical adenocarcinoma can be more difficult to Epidemiology and Risk Factors diagnose than its squamous cell counterpart. As stated previolIsly, cervical adenocarcinoma Most lesions originate in the endocervical canal seems to have increased in relative and, perhaps, and extend into the cervical os, becoming appar absolute incidence in the last 20 years. The rea ent on speculum examination only when they are on 24 September 2021 by guest. Protected copyright. sons for the suggested increase in absolute inci bulky and cause symptoms.3') Several studies re dence remain unclear. Two case-control studies port an even higher false-negative Papanicolaou have reported an association with oral contracep smear rate for adenocarcinomas than the 20 to 45 tive use (relative risk approximately 2.0 for 5 to 9 percent rates reported for squamous cell neo years of use versus no use).31i One study showed plasms . .f() In addition, some experts have sug no apparent dose-response relation, however, and gested that adenocarcinomas account for a larger no shldy has adequately addressed potential con than expected proportion of so-called rapidly founding variables such as sexual behavior. Some progressive cervical cancers, thus narrowing the of the earliest studies to document an increased window of opportunity for in situ diagnosis.41,42 incidence of cervical adenocarcinoma, using data In the present case adenocarcinoma was diag from the mid-1970s, showed that women nosed by colposcopic biopsy, while the Papanico younger than 35 years had experienced the high laou smear simultaneously showed only atypical est increase in disease incidence.2.f,27 This cohort glandular cells of uncertain distinction, favoring a included the first generation of women taking premalignant process. Findings on a Papanico-
38 ]ABFP Jan.-Feb.1997 Vol. 10 No.1 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from laou smear done several months earlier were nor Table 2. International Federation of Gynecology and mal. Based on what is known about the natural Obstetrics (FIGO) Classification of Cervical Cancer. history of cervical malignancies, the adenocarci Stage Description noma was probably present at the time of the normal findings on the Papanicolaou smear, im o Carcinoma in situ, intraepithelial carcinoma plying a false-negative test. This sequence illus Carcinoma strictly confined to cervix (extension trates why the Papanicolaou smear should not be to corpus should be disregarded) used to rule out cervical malignancy, especially in IA Microinvasive carcinoma (early stromal invasion) the face of unexplained symptoms. IE All other cases of stage I; occult cancer should be When patients with adenocarcinoma of the marked "occ" cervix are referred for colposcopy, further diffi II Carcinoma extends beyond cervix but has not culties can be encountered. No standard colpo extended to the pelvic wall; involves vagina, but not as far as lower t11ird scopic criteria exist for describing or detecting IIA No obvious parametrial involvement adenocarcinomas.43 At least one expert colpo scopist lists mistaking adenocarcinoma for a lIB Obvious parametrial involvement condyloma as a common pitfall in diagnosis. 43 III Carcinoma has extended to pelvic wall. No cancer-free space between the tumor and This discovery was especially interesting given the pelvic wall on rectal examination. Tumor the colposcopist's prediagnosis comments per involves lower third of vagina. All cases with hydronephrosis or nonfunctioning kidney taining to the appearance of the adenocarcinoma are included ("like a wart") in this case report. Routine col po IlIA No extension to pelvic wall scopic biopsy cannot differentiate AIS from inva IIIB Extends to pelvic wall or hydronephrosis or sive adenocarcinoma; at a minimum, cone biopsy nonfunctioning kidney must be performed.34 IV Carcinoma has extended beyond true pelvis or Increased recognition of potential symptoms of has clinically involved the mucosa of bladder or rectum cervical adenocarcinoma by primary care physi IVA Growili spread to adjacent organs cians might allow detection at an early stage. Whereas cervical dysplasia is usually asympto !VB Growili spread to distant organs matic, invasive cervical cancer is often sympto Adapted from Scott.49 matic. The most common symptoms are irregular vaginal bleeding and bloody discharge, but a non bloody vaginal discharge may be the only symp or during treatment reveal more advanced dis http://www.jabfm.org/ tom of cervical cancer. By pooling the great num ease. For stage IV disease, common sites of ber of case series from around the world that metastasis are lung, bone, liver, brain, peri reported initial symptoms, it is estimated that the toneum, and distant lymph nodes. 50 first symptom of 5 to 10 percent of cervical ade Survival for patients with cervical adenocarci nocarcinomas is discharge alone.7,Il,17,27,30,41,44-48 noma, as for those with squamous cell carcinoma, Thus, the importance of pursuing the diagnosis is highly dependent on clinical stage at diagno on 24 September 2021 by guest. Protected copyright. of an unexplained vaginal discharge is apparent. sis. 11 ,13,15,16,B,46 The following other factors Even when findings from naked-eye inspection might help determine prognosis, however: histo of the cervix and Papanicolaou smear are unre logic grade, tumor size, depth of invasion, lymph markable, colposcopy and endocervical curettage node status, and lymph-vascular space status. should be considered to help exclude cervical Lymph node status is particularly important, with cancer. survival from stage I and II lesions decreasing from more than 80 percent to the 10 to 50 per B Staging cent range at 5 years if nodes are positive. In Adenocarcinoma of the uterine cervix is staged contrast to squamous cell carcinoma, for which using the 1987 revision of the International Fed there is an inverse relationship between the num eration of Gynecology and Obstetrics (FIGO) ber of positive lymph nodes and survival rates, the classification for cervical cancer (Table 2).49 Al prognosis for cervical adenocarcinoma is equally though clinical staging criteria are subjective, the worsened whether one node or many nodes are clinical stage is not changed if findings at surgery involved.33
Adenocarcinoma of Cervix 39 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from
Treatment squamous cell carcinoma, but case-matched stud The approach to treatment of adenocarcinoma of ies are needed to settle this question. I) Until that the cervix is dependent on clinical staging. For time, the approach to treatment for adenocarci stage I and II disease, radical hysterectomy (in noma should be the same as for the more com cluding bilateral salpingo-oophorectomy and mon squamous cell carcinOlna. 33 pelvic lymph node dissection) is the treatment Although recurrence of tumor is most common with the highest cure rate and lowest risk of com during the first 2 years after diagnosis, recurrence plications.3.4 Whereas some studies of radiation of adenocarcinoma up to 5 years after definitive therapy for stage I and II disease have achieved therapy is not unusual, in sharp contrast to squa comparable results, most report lower success mous cell carcinoma for which recurrence after 2 rates and higher complication rates. Rectovaginal years is rare.H Frequent follow-up visits should fistula is particularly troublesome, occurring in be continued for at least 4~ months after defini up to 20 to 30 percent of patients who receive ra tive treatment of cervical adenocarcinoma. Tu diation therapy}! Although many authors advo mor can recur locally or at distant sites, but in ei cate ovariectomy to eliminate a potential site of ther case the prognosis is dismal, and almost all metastasis and recurrence, ovarian metastasis is patients with recurrent tumor eventually suc infrequently encountered in patients with stage I cumb to the disease. Median time from recur disease, and then usually only when there are pos rence to death in one review was 23.6 months itive lymph nodes. As a result, some advocate a (range 6 to 71 months), numbers typical of most more selective approach to ovariectomy. 51 series.44 Most patients with recurrence had Combination surgical and radiation therapy for deeply invasive, large-volume tumors with posi stage I and II disease is controversial. Most stud tive lymph nodes at the time of diagnosis. No ies have not found improved survival with com good salvage therapy exists. bined therapy, and the increase in posttherapy complications with radiation therapy has limited Conclusion this approach. A possible exception would be Because adenocarcinoma of the cervix has in bulky stage I and II lesions, in which preoperative creased in relative and possibly absolute inci radiotherapy could be beneficial. l'J,3),H dence, it is important for family physicians to un Patients with stage III and IV disease fare derstand its natural history, particularly as it poorly. Radiotherapy and chemotherapy are gen differs from squamous cell cancer. Using screen erally offered as first-line treatments, with sur ing Papanicolaou smears, the premalignant phase http://www.jabfm.org/ gery reserved for debulking and palliation.33 of adenocarcinoma, AIS, can be more difficult to diagnose than squamous dysplasia or carcinoma Prognosis in situ preceding invasive squamous cell cancer. The prognosis for adenocarcinoma of the uterine This difficulty reflects the tendency of adenocar cervix is primarily dependent on the FTGO stage cinoma to originate within the cervical os, out of at diagnosis. Overall 5 -year survival rates in ex reach of a spatula and even a cytologic sampling on 24 September 2021 by guest. Protected copyright. cess of ~o percent are consistently obtained for brush, until it becomes bulky or symptomatic. surgically treated stage I disease. The 5-year The lack of a characteristic appearance at col prognosis is uniformly dismal with increasing poscopy also contributes to diagnostic difficulty. FIGO stage, however, decreasing to 20 to 30 per Despite these pitfalls, early diagnosis is essential if cent for stage III and approximately 5 percent for successful treatment is to be initiated. As is squa stage IV35 Many authors report decreased sur mous cell cancer, adenocarcinoma of the cervix is vival for all stages of cervical adenocarcinoma curable in more than ~o percent of cases, usually compared with cervical squamous cell carcinoma by surgery, when detected at stage I. The progno of equivalent stage, but they used unmatched pa sis for advanced-stage tumors, however, remains tients treated at the same or other institutions as a dismal no matter which therapy is chosen. Pri control group, a serious methodologic flaw that mary care research into better screening and di limits the strength of their conclusions. 40 ]ABFP Jan.-Feb. 1997 Vol. 10 No.1 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from noma. Less clear is the intriguing assertion that cell cancer and adenocarcinoma of the uterine oral contraceptive use might be a risk factor, cervix. Gynecol OncoI1989;33:340-3. which could account for the increasing number of 10. Duk ]M, Aalders ]G, Fleuren G], Krans M, De Bruijn HW Tumor markers CA 125, squamous cell cases since the mid-1970s in women younger carcinoma antigen, and carcinoembryonic antigen in than 35 years, 15 years after widespread usage of patients with adenocarcinoma of the uterine cervix. birth control pills began. While this theory is cur Obstet GynecoI1989;73:661-8. rently supported only by observational, popula 11. Eifel P], Morris M, Oswald M], Wharton ]T, tion-based studies, the consistency of reported ef Delclos L. Adenocarcinoma of the uterine cervix. fect from multiple sites worldwide suggests that Prognosis and patterns of failure in 367 cases. Can cer 1990;65:2507-14.. case-control and cohort studies in the primary 12. Eide T]. Cancer of the uterine cervix in Norway by care setting are warranted. histologic type, 1970-84.] Nat! Cancer Inst 1987; Not only does the case in this report illustrate 79:199-205. several of the distinctive features of cervical ade 13. Goodman HM, Buttlar CA, Niloff]M, Welch WR, nocarcinoma, it also provides another clear mes Marek A, Feuer E], et al. Adenocarcinoma of the sage to family physicians: using the Papanicolaou uterine cervix: prognostic factors and patterns of re smear to rule out invasive cervical cancer is ill currence. Gynecol OncoI1989;33:241-7. 14. Hopkins MP, Morley Gw. A comparison of adeno advised. In the face of unexplained vaginal bleed carcinoma and squamous cell carcinoma of the ing or discharge or a grossly visible cervical le cervix. Obstet GynecoI1991;77:912-7. sion, the diagnosis of cancer must be entertained 15. Kilgore LC, Soong S], Gore H, Shingleton HM, even if the results of a Papanicolaou smear are Hatch KD, Partridge EE. Analysis of prognostic normal. By maintaining a high index of suspicion features in adenocarcinoma of the cervix. Gynecol for cervical cancer in the face of suggestive symp OncoI1988;31:137-53. toms, an early diagnosis and an excellent progno 16. Kleine W, Rau K, Schwoeorer D, Pfeiderer A. Prog nosis of adenocarcinoma of the cervix uteri: a com sis for the patient can be achieved. parative study. Gynecol Oncol 1989;35: 145-9. 17. Leminen A, Paavonen], Forss M, Wahlstrom T, References Vesterinen E. Adenocarcinoma of the uterine cervix. 1. Angel C, DuBeshter B, LinJY. Clinical presentation Cancer 1990;65:53-9. and management of stage I cervical adenocarcinoma: 18. Milsom I, Friberg LG. Primary adenocarcinoma of a 25 year experience. Gynecol OncoI1992;44:71-8. the uterine cervix: a clinical study. Cancer 1983;52: 2. Anton-Culver H, Bloss]D, Bringman D, Lee-Feld 942-7. stein A, DiSaia P, Manetta A. Comparison of adeno 19. Moberg P], Einhorn N, Silfversward C, Soderberg carcinoma and squamous cell carcinoma of the uter G. Adenocarcinoma of the uterine cervix. Cancer http://www.jabfm.org/ ine cervix: a population-based epidemiologic study. 1986;57:407-10. Am] Obstet GynecoI1992;166:1507-14. 20. Parazzini F, LaVecchia C, Negri E, Fasoli M, Cec 3. Berek]S, Castaldo TW, Hacker NF, Petrilli ES, La chetti G. Risk factors for adenocarcinoma of the gasse LD, Moore]G. Adenocarcinoma of the uter cervix: a case-control study. Br] Cancer 1988;57: ine cervix. Cancer 1981;48:2734-41. 201-4. 4. Berkowitz RS, Ehrmann RL, Lavizzo-Mourey R, 21. Peters RK, Chao A, Mack TM, Thomas D, Bern Knapp Re. Invasive cervical carcinoma in young stein L, Henderson BE. Increased frequency of ade on 24 September 2021 by guest. Protected copyright. women. Gynecol OncoI1979;8:311-6. nocarcinoma of the uterine cervix in young women in Los Angeles County.] Nat! Cancer Inst 1986;76: 5. Brinton LA, Herrero R, Reeves WC, deBritton RC, 423-8. Gaitan E, Tenorio F. Risk factors for cervical cancer 22. Raju KS, Bates TD, Taylor RW. Primary adenocar by histology. Gynecol OncoI1993;51:301-6. cinoma of the cervix: treatment and results. Br J Ob 6. Chilvers C, Mant D, Pike Me. Cervical adenocarci stet Gynecol 1987;94: 1212 -7. noma and oral contraceptives. Br Med] Clin Res Ed 23. Schwartz SM, Weiss NS. Increased incidence of 1987;295:1446-7. adenocarcinoma of the cervix in young women 7. Davis ]R, Moon LB. Increased incidence of adeno in the United States. Am] Epidemiol 1986;124: carcinoma of uterine cervix. 0 bstet Gynecol 1975; 1045-7. 45:79-83. 24. Shingleton HM, Gore H, Bradley DH, Soong S]. 8. Devesa SS, Young]L]r, Brinton LA, Fraumeni]F Adenocarcinoma of the cervix: I. Clinical evaluation Jr. Recent trends in cervix uteri cancer. Cancer and pathologic features. Obstet GynecoI1981;139: 1989;64:2184-90. 799-814. 9. Drescher Cw, Hopkins MP, Roberts]A. Compari 25. Silcocks PB, Thornton-]ones H, Murphy M. Squa son of the pattern of metastatic spread of squamous mous and adenocarcinoma of the uterine cervix: a Adenocarcinoma of Cervix 41 J Am Board Fam Pract: first published as 10.3122/jabfm.10.1.36 on 1 January 1997. Downloaded from comparison using routinc dat,). Br .I Cancer tomic distribution of cervical adenocarcinoma in 19H7;55:321-5. situ: implications for treatment. Am j Obstet Gyne Hi. Stegner I fE. A growing concern: cervical adenocar coI19H7;157:21-5. cinoma (meeting abstract). Presented at the Interna 40. Ng'uyen GK, Daya D. Cervical adenocm:inom status of I 14 endocervical adenocarcinoma cases by 51. Brown jY, Fu YS, BerekjS. Ovarian metastases are on 24 September 2021 by guest. Protected copyright. dot blot hybridization. Hum PathoI1993;24: 121-5. rare in stage I adenocarcinoma of the cervix. Obstet 39. Bertrand M, Lickrish GM, Colgan TJ. The ana- Gynecol 1990; 7 6:623 -6. 42 JABFP jan.-Feb.1997 Vol. 10 No. I