Annals of C linical and Laboratory Science, Vol. 4 , No. 6 Copyright © 1 9 7 4 , Institute for Clinical Science

Recent Developments in the Cytologic Diagnosis of Y aginal, Endometrial and Ovarian Cancer

BERNARD GONDOS, M.D.

Department of Pathology, University of California, San Francisco, CA 94143

ABSTRACT Cytology has proved to be an effective and simple technique for the di­ agnosis of cancer of the uterine . The usefulness of cytologic methods in detecting other types of tumors of the female genital tract has been rela­ tively limited. In recent years, new developments have focused attention on the use of cytology in the following areas: (1) detection of adenosis and of the vagina in daughters of women who received synthetic during pregnancy; (2) detection of early stages of adenocarcinoma of the by direct sampling methods and (3) prognostic evalua­ tion of cancer of the ovary by examination of peritoneal washings. The present report reviews the cytologic methods involved in these procedures with consideration of both technical and interpretive aspects.

Introduction that traditional cytologic methods have Cytologic diagnosis of cancer of the been ineffective in preventing the rise. uterine cervix has achieved great success Similarly, cancer of the ovary remains a over the past 25 years. Detection of cer­ difficult diagnostic problem. The inacces­ vical neoplasia in the stages of dysplasia, sible location of the ovaries makes cyto­ carcinoma in situ and early invasive car­ logic sampling particularly difficult and cinoma has enabled treatment and cure of new approaches in this area are badly a potentially fatal disease in numerous needed. cases. Evidence of this is the marked de­ Cytologic diagnosis of vaginal cancer cline in mortality from cancer of the cervix presents a different type of problem. Clari­ in the past two decades. fication is required on the role of cytology Unfortunately, the same cannot be said in regard to the newly recognized condi­ for cancer of the endometrium and ovary, tions of adenosis and adenocarcinoma of the two other common sites of malignancy the vagina occurring in young women in the female genital tract. Endometrial whose mothers received synthetic estro­ cancer has shown a steady increase in in­ gens during pregnancy. Because of the ex­ cidence in recent years. Although the tremely high number of possibly affected reasons for this are not known, it is evident individuals and the serious implications in­ 420 VAGINAL, ENDOMETRIAL AND OVARIAN CYTOLOGY 4 2 1

volved, early recognition of these lesions is carcinoma supports such a relationship. of special concern. The possible contribu­ Adenosis is characterized by the presence tion of cytology as a means of early detec­ of regular single-layered glandular struc­ tion is an important consideration. tures located beneath the stratified squa­ mous mucosa. The process may extend to Cancer of the Vagina the mucosal surface. The glands are gen­ Primary carcinoma of the vagina is an erally lined by tall columnar endocervical- uncommon tumor and one which, until re­ type epithelium with basally situated nu­ cently, had been found to occur predom­ clei and abundant cytoplasm (figure 1). inantly in the sixth and seventh decades in These cells are usually PAS-positive. The the form of squamous cell carcinoma. In cytologic diagnosis of adenosis is indicated 1970, Herbst and Scully16 described seven by the presence of cells resembling endo- cases of adenocarcinoma of the vagina oc­ cervical epithelium or by the presence of curring in adolescents. Detailed investiga­ endometrial-type cells. Vaginal adenosis tion in these cases suggested an association can be detected by direct scraping of with maternal administration of stilbestrol lesions which appear as reddened granular and other synthetic estrogens. This associa­ areas, most often found in the upper ante­ tion was further documented as additional rior vaginal wall. In the absence of a visi­ cases were collected.12’17’32 It thus became ble lesion, occult adenosis may be detected clear that detection of adenocarcinoma of by circumferential scraping of the vaginal the vagina in daughters of women who mucosa. In doing this, it is important to received synthetic estrogens during preg­ minimize the possibility of contamination nancy represents an important diagnostic from the endocervix by removing any se­ problem. cretion overlying the vaginal mucosa be­ Early studies indicated that routine cyto­ fore performing the scraping. The presence logic evaluation was not reliable in the of endocervical-type cells can then be diagnosis of vaginal adenocarcinoma. Of taken as evidence of vaginal adenosis. 30 cases in which specimens were obtained Adenocarcinoma of the vagina appears for cytology, 11 were positive, nine nega­ as a grossly evident tumor usually, but not tive and 10 doubtful.15 It was subsequently always, located in the upper anterior va­ reported by Vooijs et al33 that cytology was gina. The size is variable, ranging from one useful in the detection of vaginal adeno­ to ten cm in diameter. Histologic appear­ carcinoma and its precursor, adenosis, ance has been reviewed in detail else­ when direct vaginal scrapings were taken. where.16 The principal features are a well These studies indicate that the manner in differentiated glandular pattern with clear which specimens are obtained is critical. cells lining the gland spaces (figure 2). Routine sampling of cervix and vaginal The pattern may be solid in some cases pool will fail to detect many cases of ad­ but is most frequently cribriform or “hob­ enosis and adenocarcinoma, while direct nail” type. At higher magnifications, the sampling of vaginal lesions can be of di­ cells can be seen to have slightly irreg­ agnostic value. ular nuclei and large nucleoli (figure 3A). Recognition of vaginal adenosis is espe­ Cytologic diagnosis depends on direct cially important because of its probable sampling of the tumor. The cells exhibit role in the pathogenesis of adenocarci­ characteristic features of adenocarcinoma, noma.15 Although a direct relationship including aggregation in groups, increased remains to be established, the frequent nucleocytoplasmic ratio and marked nu­ coexistence of adenosis in cases of adeno­ cleolar enlargement (figure 3R). Slight to 422 f » n esn X400. eosin, and rmnn ncel, lgt ula vraiiy rltvl hmgnos hoai pten and with pattern groups chromatin cell show homogeneous preparations relatively cytologic variability, and nuclear histologic slight Both nucleoli, prominent pregnancy. during bestrol la t fnl gaua ctpam A Tsu scin Hmtxln n esn X0. . Cyto­ B. X400. eosin, and Hematoxylin X500. section; stain, Tissue Papanicolaou A. smear; cytoplasm. logic granular finely to clear e r u g i F e r u g i F e r u g i F . aia aeoi wt caatrsi edcria-ye pteim Hematoxylin epithelium. endocervical-type characteristic with adenosis Vaginal 1. . aia aeoacnm i a 5ya-l gr woe ohr a rcie stil- received had mother whose girl 15-year-old a in adenocarcinoma Vaginal 3. . aia aeoacnm, o-al atr. eaoyi ad oi, X80. eosin, and Hematoxylin pattern. hob-nail adenocarcinoma, Vaginal 2. GONDOS

VAGINAL, ENDOMETRIAL AND OVARIAN CYTOLOGY 423

T A B L E I

C y t o l o g i c D i a g n o s i s o f A d e n o s i s a n d A denocarcinoma o f t h e V a g i n a

Specimen S ta in F in d in g s

Obtain scrapings from lesion; Routine Pap Strips or sheets of benign if no lesion present, scrape stain columnar cells resembling vaginal mucosa circumferen- (PAS may be endocervical or endometrial tially, avoiding cervical helpful) cells; basal nuclei, vacuo­ contamination. lated cytoplasm, occas. Ciliated, may be PAS positive.

Adenocarcinoma Obtain scrapings from lesion. Routine Pap Groups of neoplastic cells stain with pleomorphic round to oval nuclei, multiple irregular macronucleoli, finely granular faintly eosinophilic cytoplasm.

moderate nuclear variability is present, but not being achieved. Menopausal and post­ markedly anaplastic cells are not seen. menopausal women are the prime suspects, Cytoplasm may be clear or finely granular, but cases in women under 40 are being contrasting with the hyperchromatic nu­ increasingly reported.20 In addition, there clei. The cytologic findings associated with is a general rise in the most aggressive vaginal adenocarcinoma are distinctly dif­ form, adenosquamous carcinoma.2’29 ferent from those found with adenosis as The traditional method for cytologic di­ listed in table I. agnosis of endometrial lesions is sampling Since the diagnosis of vaginal adenocar­ of the vaginal pool. This method relies on cinoma by cytologic examination has thus exfoliation of cells from the endometrial far been made only in clinically evident le­ lining into the region of the vaginal fornix. sions, it remains to be established whether Because this is an indirect form of sam­ or not a preclinical stage of adenocarci­ pling, the false negative rate is high, espe­ noma can be detected. In the meantime, it cially so for early lesions.25 The endome­ is recommended that vaginal scrapings be trial cells encountered in vaginal pool employed routinely in the examination of material are often in a poor state of pres­ adolescents whose mothers received stil- ervation; differentiation of degenerated en­ bestrol or other synthetic estrogens during dometrial cells, histiocytes, distorted endo­ pregnancy. Cytology can also be used in cervical cells and cells from endometrial following patients with a diagnosis of vagi­ hyperplasia and adenocarcinoma may be nal adenosis. Until the extent of the rela­ difficult. As a result, routine screening has tionship between adenosis and adenocarci­ contributed only rarely to the diagnosis of noma is more fully known, careful cyto­ endometrial carcinoma. logic follow-up will remain a principal A variety of methods for direct sampling means of managing patients with adenosis. of the endometrium have been utilized in an attempt to overcome these problems. The techniques involve aspiration, brush­ Cancer of the Endometrium ing or lavage. Comparison of the advan­ Endometrial cancer has shown a sharp tages and disadvantages of the different rise in recent years.6 Although improved methods is shown in table II. detection may account in part for the in­ Endometrial aspiration is the simplest of creasing incidence, failure to reduce the the direct sampling techniques. A cannula mortality indicates that early detection is is inserted into the endometrial cavity and 4 2 4 GONDOS

T A B L E II cells into the peritoneal cavity via the C o m p a r i s o n o f E n d o m e t r i a l fallopian tubes. There may also be con­ C y t o l o g i c T e c h n i q u e s siderable difficulty in interpretation result­

Method Advantages Disadvantages ing in a significant false positive rate. The accuracy of different lavage techniques has Vaginal Simple; Indirect sampling been reported to range from 80 to 90 per­ pool inexpensive. depending on cell­ ular exfoliation; cent.24’31 high false negative rate; difficult to Recently, a new endometrial lavage tech­ interpret. nique employing the Gravlee jet washer* Aspiration Simplest of Specimen may be has been recommended as an effective the direct bloody; possibility methods; good of perforation. screening technique with a high degree of cellular yield; good accuracy.8’30 This method utilizes a dis­ cell detail. posable double lumen cannula, through Brushing Good cellular Expensive ; involved yield. procedure; specimen which 30 to 40 ml of saline are washed to may be bloody; collect material in a plastic container. The possibility of infection. application of suction by a syringe in one

Lavage Good cellular High false positive tube creates a vacuum, exerting negative yield. rate in early reports; involved pressure and thereby avoiding spread of procedure; expensive; malignant cells. With this technique, possible spread of malignant cells. smears, millipore filters and cell blocks can be prepared, the latter frequently contain­ ing tissue fragments. The procedure is in­ the aspirated material is centrifuged to volved and expensive, but has received form a cell button from which smears and considerable use. Reports on results are histologic sections can be prepared. The conflicting. Some have been enthusiastic accuracy of aspiration has ranged from 70 about the lack of technical problems and to 90 percent in different studies.14’18’19’84 high detection rate,4’35 while others have Major disadvantages include the risk of found a considerable number of specimens perforation, the frequent admixture of ex­ to be unsatisfactory, particularly in post­ cessive blood and the requirement that the menopausal women.1’21 There is general procedure be performed by a specialist. agreement that the technique compares Endometrial brushing, employing a vari­ favorably with endometrial biopsy in the ety of brushes for obtaining direct smears, presence of abundant neoplastic tissue, but has an accuracy of 57 to 92 percent in cases its use as a screening procedure in asymp­ of histologically confirmed adenocarci­ tomatic women has not been universally noma.3’9’13’19 The technique is too involved supported.28 and expensive for routine use and has the Of all the methods for direct endome­ additional drawback of possible introduc­ trial sampling, endocervical aspiration ap­ tion of infection. pears to be the simplest and the one most Lavage techniques involve irrigation of widely employed.27’34 This technique can the uterine cavity with saline and collec­ be utilized as part of the gynecologic ex­ tion of the fluid in centrifuge tubes. After amination in conjunction with routine the material is centrifuged, smears and cell screening for cervical disease. Ng et al26 blocks can be prepared from the sediment. have recently described the use of endo­ Various types of cannulas have been used cervical aspirates to evaluate cytologic to collect specimens. All of the techniques have the potential risk of forcing malignant * Upjohn Company, Kalamazoo, MI 49001. VAGINAL, ENDOMETRIAL AND OVARIAN CYTOLOGY 4 2 5 criteria for diagnosing precursors of endo­ rial. Satisfactory specimens were obtained metrial cancer, such as adenomatous hy­ in only 70 percent of aspirations, and in perplasia, atypical hyperplasia and adeno­ many others the cellular yield was minimal. carcinoma in situ. They were able to Furthermore, the question has been establish satisfactory criteria for distin­ raised as to whether or not it is too late guishing these conditions from endometrial for effective therapy when malignant cells carcinoma, although agreeing with others are already present in the cul-de-sac wash­ that even with the aspiration technique ings. Graham11 has expressed the opinion interpretation is difficult and requires con­ that the presence of malignant cells in the siderable experience. The technique clearly aspirate does not necessarily mean ad­ provides a more constant and greater con­ vanced disease, since the cells might be centration of endometrial cells than tradi­ exfoliated without implantation on the per­ tional indirect sampling techniques and, itoneal surface. On the other hand, many with increasing experience, may prove to of the common ovarian tumors of the sur­ be effective in detecting precursors of en­ face epithelium remain confined within the dometrial cancer. ovarian parenchyma during their early growth and exfoliation into the peritoneal cavity would be a late event in these cases. Cancer of the Ovary Peritoneal washings obtained at laparot­ The overall survival rate for ovarian can­ omy have also been used in the diagnosis cer has not changed appreciably in over a of ovarian cancer. Creasman and Rutledge7 quarter of a century, largely owing to late reported on a ten year experience with diagnosis. Attempts to utilize cytologic de­ more than 1,200 specimens. Most were ob­ tection have been ineffective. Again, the tained in cases of known ovarian cancer problem is one of providing a direct for prognostic evaluation. Also included method of sampling that is simple and was a group of 112 patients who were reliable. operated on owing to benign gynecologic Culdocentesis, or cul-de-sac aspiration, disease. One hundred to 150 cc of saline has been used as a method for detecting were instilled into the peritoneal cavity cells from ovarian tumors over the past and allowed to collect in the pelvic region. decade. In 1964, Graham et al10 reported The fluid was then removed and treated on a series of 516 cul-de-sac aspirations in the usual manner for cytologic study, obtained from asymptomatic volunteers. smears and cell blocks being prepared There were eight positive cases, of which from the spun-down sediment. Abnormal seven were explored: one had metastatic cells were reported in five of the asympto­ breast cancer, four had papillary lesions in matic patients, none of whom was found the ovary of “borderline malignancy,” one to have malignant disease. However, three had a “probable borderline lesion” and one had ovarian cystadenomas and two had had no demonstrable lesion. In reviewing benign conditions involving the . these findings, Koss22 suggested that the These results represent a significant false method of culdocentesis required further positive rate and point out the hazards of evaluation and should be given a wide interpretation in peritoneal washings. In field trial. Subsequently, over 1,400 asymp­ the patients with known ovarian cancer, tomatic women were screened in two sepa­ those with abnormal cytologic findings had rate studies,5’23 with only one positive di­ survival rates considerably lower than agnosis. A serious problem of the method those with normal findings. Cytology was was in obtaining sufficient cellular mate­ therefore considered useful in prognostic 4 2 6 GONDOS assessment and evaluation of further treat­ 15. H erbst, A. L., K urman, R. J., Scully, R. E ., ment. and Poskanzer, D. C.: Clear-cell adenocar­ cinoma of the genital tract in young females. For the present, it appears that cytology New Eng. J. Med. 287:1259-1264, 1972. has little to offer in the initial diagnosis of 16. Herbst, A. L. and Scully, R. E.: Adenocar­ ovarian cancer, but it can be utilized for cinoma of the vagina in adolescence. Cancer 25:745-757, 1970. the purpose of prognostic evaluation 17. Herbst, A. L., Ulfelder, H., and Poskan­ through the use of peritoneal washings. zer, D. C.: Adenocarcinoma of the vagina: Association of maternal stilbestrol therapy with tumor appearance in young women. New R eferences Eng. J. Med. 284:878-881, 1971. 1. A b a t e , S. D., Creighton, L. E., a n d V e l - 18. Isaacs, J. H. and Wilhoite, R. W .: Aspira­ l i o s , F .: A comparative study of the endome­ tion cytology of the endometrium: Office and trial jet-washing technic and endometrial bi­ hospital sampling procedures. Amer. J. Obstet. opsy. Amer. J. Clin. Path. 58:118-122, 1972. Gynec. 118:679-684, 1974. 2. A i k a w a , M. and Ng, A. B. P.: Mixed (adeno- 19. J ameson, M. H .: A clinical appraisal of tech­ squamous) carcinoma of the endometrium: niques for obtaining fresh cells for endome­ Electron microscopic observations. Cancer 31: trial cytology. New Zeal. Med. J. 60:316-332, 385-397, 1973. 1961. 3. A y r e , E. T.: Rotating endometrial brush: 20. Kempson, R. L. and Pokorny, G. E .: Adeno­ New technic for the diagnosis of fundal car­ carcinoma of the endometrium in women cinoma. Obstet. Gynec. 5:137-141, 1955. aged forty and younger. Cancer 21:650-662, 4. B i b b o , M., S h a n k l i n , D. R., a n d W i e d , G. 1968. L.: Endometrial cytology on jet wash mate­ 21. Klin e, T. S. and Ring, I. R.: Evaluation of rial. J. Reprod. Med. 8:90-96, 1972. endometrial cytologic techniques. Lab. Med. 5. Bolandgray, A., M ahallati, K. A., and 4:27-28, 1973. Ardekany, M. S.: Early detection of ovarian 22. Koss, L. G.: Diagnostic Cytology and Its malignancy by culdocentesis. J. Reprod. Med. Histopathologic Bases. J. B. Lippincott Co., 9:32-34, 1972. Philadelphia, 2nd ed., pp. 268-269, 1968. 6. Christopherson, W . M., Mendez, W . M., 23. McG owan, L., Stein, D. B., and Miller , P a r k e r , J. E., L u n d e n , F. E., a n d A h u j a , W.: Cul-de-sac aspiration for diagnostic cyto­ E . M . : Carcinoma of endometrium: A study logic study. Amer. J. Obstet. Gynec. 96:413- of changing rates over a 15-year period. Can­ 417, 1966. cer 27:1005-1008, 1971. 24. Morton, D. G., Moore, J. G., and Chang, 7. C r e a s m a n , W. T. and Rutledge, F .: The N.: Lavage of the endometrial cavity as an prognostic value of peritoneal cytology in aid in the diagnosis of carcinoma of the uter­ gynecologic malignant disease. Amer. J. Ob­ ine corpus. J. Intl. Coll. Surg. 31:570-578, stet. Gynec. 110:773-781, 1971. 1959. 8. D o w l i n g , E. A., G r a v l e e , L. C., a n d 25. Nahhas, W. A., Curtis, J. L., and Rudolph, H u t c h i n s , K. E.: A new technique for the J. H.: Carcinoma of the corpus uteri. A 10- detection of adenocarcinoma of the endome­ year review of 225 patients. Obstet. Gynec. trium. Acta Cytol. 13:496-501, 1969. 38:564-570, 1971. 9. Fox, C. H., T u r n e r , F. G., J o h n s o n , W. L „ 26. Ng, A. B. P., Reagan, J. W., and Cechner, and Thornton, W. N.: Endometrial cytol­ R. L.: The precursors of endometrial cancer: ogy, a new technique. Amer. J . Obstet. A study of their cellular manifestations. Acta Gynec. 83:1582-1591, 1962. Cytol. 17:439-448, 1973. 10. G r a h a m , J. B., G r a h a m , R. M., and Schuel- 27. Reagan, J. W. and Ng, A. B. P.: The Cells l e r , E. F.: Preclinical detection of ovarian of Uterine Adenocarcinoma. Williams & Wil­ cancer. Cancer 77:1414-1432, 1964. kins, Baltimore, 1965. 11. G r a h a m , R. M.: Diagnosis of ovarian carci­ 28. Rodrigues, M. A., Rubin, A., Koss, L. G., noma by cul-de-sac aspiration. Ovarian Can­ and H arris, J.: Evaluation of endometrial jet cer. Gentil, F. and Junqueira, A. C., eds., wash technic (Gravlee) in 303 patients in a Springer-Verlag, Berlin, pp. 122-133, 1968. community hospital. Obstet. Gynec. 43:392- 12. Greenwald, P., B a r l o w , J. J., N a s c a , P. C., 399, 1974. and Burnett, W. S.: Vaginal cancer after 29. Silverberg, S. G., B olin, M. G., and D e - maternal treatment with synthetic estrogens. Giorgi, L. S.: Adenoacanthoma and mixed New Eng. J. Med. 285:39(^-392, 1971. adenosquamous carcinoma of the endome­ 13. H a m r o v a , D., Laurova, L., Skoda, V., a n d trium. Cancer 30:1307-1314, 1972. T r n k a , V.: Brush technic for endometrial 30. So-Bosita, J. L., L ebherz, T. B., and Blair, cytologic study. Neoplasma 19:239-242, 1972. O. M.: Endometrial jet washer. Obstet. 14. H e c h t , E. L.: The endometrial aspiration Gynec. 36:287-293, 1970. smear: Research status and clinical value. 31. T orres, J. E., Holmquist, N. D., and Amer. J. Obstet. Gynec. 71:819-833, 1956. D anos, M. L .: The endometrial irrigation VAGINAL, ENDOMETRIAL AND OVARIAN CYTOLOGY 4 2 7

smear in the detection of adenocarcinoma of clear cell carcinoma. Acta Cytol. 17:59-63, the endometrium. Acta Cytol. 13:163-168, 1973. 1969. 34. W a c h t e l , E., G o r d o n , H., a n d W y c h e r l e y , 32. T s u k a d a , Y., H e w e t t , W. J., B a r l o w , J. J., J.: The cytologic diagnosis of endometrial a n d P i c k r e n , J. W.: Clear-cell adenocarci­ pathology using a uterine aspiration tech­ noma (“mesonephroma”) of the vagina. nique. J. Obstet. Gynec. Brit. Comm. 80:164- Three cases associated with maternal syn­ 168, 1973. thetic nonsteroid therapy. Cancer 29: 35. W h i t e , A. J., R o d m a n , N. F., a n d B u c h s - 1208-1214, 1972. b a u m , H. J.: Clinical use of the Gravlee jet 3 3 . V o o i j s , P. G ., N g , A. B . P., a n d W e n t z , washer. J. Iowa Med. Soc., pp. 211-215, May W. B.: The detection of vaginal adenosis and 1973. Manual of Procedures for the Applied Seminar on the Laboratory Diagnosis of Skeletal, Muscular and Nervous Disorders The following procedures are included in this listing: Direct Measurement of Serum Calcium Direct Measurement of Serum Inorganic Phosphate Practical Aspects of Radioimmunoassay of Parathyroid Hormone Radioimmunoassay for Human Parathyroid Hormone Measurement of Alkaline Phosphatase in Serum Determination of Serum Alkaline Phosphatase Isoenzymes Measurement of CPK Isoenzymes in Serum Hypoxanthine-guanine Phosphoribosyltransferase ( HGPRT ) Assay Detection of Tay-Sachs Heterozygotes by Determination of /3-N-Acetyl-D-Hexosa- minidase A Measurement of Blood Lactate A Kinetic Colorimetric Method for the Measurement of Uric Acid in Serum A Kinetic Uricase Method for Measurement of Uric Acid Measurement of Salicylates in Serum Measurement of Serum Gold in the Regulation of Chrysotherapy of Rheumatoid Arthritis Measurement of Folic Acid and Vitamin B12 in Seram Measurement of Homovanillic Acid in Cerebrospinal Fluid Latex and Sheep Cell Agglutination of Rheumatoid Factor Methods in the Immunofluorescent Diagnosis of Connective Tissue Diseases Streptococcal Antibody Tests (Antistreptolysin O, Antihyaluronidase, AntiDNase B, Antistreptokinase, AntiNADase ) Electroimmuno Assay Procedures for the Measurement of Orosomucoid, Alpha-1- Antitrypsin and Ceruloplasmin Evaluation of Acute Phase Reactants The Fluorescent Antibody Test for Toxoplasmosis Glutamine and Ammonia Measurements in Cerebrospinal Fluid The Laboratory Diagnosis of Galactosemia Measurement of Acid Mucopolysaccharides Measurement of Serum Dilantin and Other Antiepileptic Drugs Measurement of Serum Lithium in the Regulation of Lithium Therapy Protein Measurements in Cerebrospinal Fluid Polyacrylamide Gel Electrophoresis of Spinal Fluid Proteins Measurement of Lead in Biological Fluid Sensitive Determination of Urinary Coproporphyrin and Uroporphyrin by Chroma­ tography and Fluorometry Thin Layer Chromatography for Urinary Porphyrins Serologic Tests for Viral Myocarditis Needle Biopsy of Bone Marrow Published by the Institute for Clinical Science, Inc. for the Association of Clinical Scientists Special Price $13.00 Order directly from: Institute for Clinical Science, Inc., F. William Sunderman, M.D., Director, 1833 Delancey Place, Philadelphia, Pennsylvania 19103