Assessment of Results of Estrogen and Progesterone Receptor Assays Performed in a Community Hospital

ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 16, No. 4 Copyright 1986, Institute for Clinical Science, Inc. Assessment of Results of Estrogen and Progesterone Receptor Assays Performed in a Community Hospital JULIANA G. SZAKACS, M.D.,* JORGE G. ARROYO, M.D.,tl anda N TH O N Y J. GIRGENTI, P h.d. *Department of Family Medicine, Beaver County Hospital, Beaver, PA 15009 and f Department of Pathology and Department of Laboratories, St. Joseph s Hospital, tuniversity of South Florida, College of Medicine, Tampa, FL 33607 ABSTRACT A pproxim ately 1,000 assays for estrogen recepto r (ER) in prim ary hum an breast tum ors have been perform ed at St. Joseph s Hospital over a period of seven years; 700 of these included assays for progesterone receptor (PR). Based on the m ethod of analysis (dextran-coated charcoal) and criteria for a positive result used for this survey, 80 percent of the prim ary tum ors w ere ER-positive and 56 percent were PR-positive. In those cases w here both assays w ere perform ed, 47.4 percent were ERpositive, PR-positve; 19.8 percent w ere ER-positive, PR-negative; 6.2 percent w ere ER-negative, PR-positive; and 26.6 percent were ER-negative, PR-negative. The m ean concentration of ER increased with the advancing age of th e patient; essentially th e same relationship was observed for PR. The concentration of ER and PR was not directly dependent upon the degree of cellularity of the tumor. Lobular carcinoma and the mixed types containing ductal and lobular elem ents had the highest frequency of being positive for both steroid receptors, while m edullary and papillary carcinomas were lowest. Three hundred and twenty-two cases had follow-up studies and w ere exam ined on the basis of the available inform ation in the files of St. Joseph s Hospital Tumor Registry. A higher survival rate in patients with both ER and PR positivity became evident. In a com m unity hospital setting, our data confirm the usefulness of estrogen and progesterone receptor assays in decisions of clinical management and considerations of prognosis in patients with mammary carcinoma. Introduction The treatm en t of breast cancer, the m ost com m on m alignancy in w hite females over 40 years of age, is a subject receiving m uch attention, investigation, and change in opinion. Among the different surgical alternatives, m odified 266 0091-7370/86/0700-0266 $01.20 Institute for Clinical Science, Inc.

radical m astectom y rem ains the p re ferred treatm ent. Recurrences and dissem inated tum ors are treated by radiotherapy, cytotoxic chem otherapy, and hormonal manipulation. For m any years it was know n that removal of the ovaries, adrenals or the pituitary had a beneficial effect in some cases of carcinom a of the b reast.3 The sam e effect was o b tain ed using high doses of estrogen near the toxic range. Recently, com pounds almost devoid of side effects, called anti-estrogens, have becom e available for horm one sensitive tum ors.7 Even though endocrine therapy appears to benefit only one-third of the cases, the quality of their remaining life seems to be appreciably better.7 A major advantage came about when it was p o ssib le to p re d ic t w ith a high degree of accuracy the hormonal receptor status of each tum or.9 M ethods for cytoplasmic estrogen and progesterone receptors w ere introduced, and correlative studies led to the application of the data for clinical decisions in the managem ent of cancer patients. The flow chart, figure 1, widely used in clinical practice for the treatm ent of breast cancer, indicates the im portance of the laboratory reports in making therapeutic decisions. Regardless of the mass of the tumor, endocrine therapy is used first in those cases which contain hormonal receptors. In those patients whose receptor status precludes the use of horm onal tr e a tm e n t, c h e m o th e ra p y is started immediately.13 DISSEMINATED DISEASE / 1 \ ER+/PR+ ER+/PR- ER-/PRi I I E ndocrine A nti estrogens Chemotherapy Therapy + Chemotherapy Chemotherapy F IG U R E 1. F l o w c h a r t o f c l i n i c a l d e c i s i o n s i n m a n a g e m e n t o f b r e a s t c a n c e r. A S S E S S M E N T O F E R A N D P R IN B R E A S T C A N C E R 2 6 7 Since 1976, our laboratory has ro u tinely p erfo rm ed assays for estro g en receptors in breast tum or tissue. Reginning in 1979, assays have also been perform ed for receptors to progesterone, since it has been shown that the predictive value is enhanced when both assays are considered together.9,12,18 The analytical m ethods for estrogen and progesterone receptor are relatively complicated and require a great deal of handling. The quality control data of these m ethods may show a wide range of v aria b ility. C o n tro l tissu e has only recently been made commercially available.13 Under these circumstances especially, the clinical utility of these assays must be closely evaluated. T he Tum or R egistry has follow -up records in 322 cases of breast carcinoma diagnosed and treated at St. Jo seph s Hospital (SJH) during a four year period. These records w ere used to derive survival statistics in relation to estrogen and progesterone receptor status. M aterials and Methods The m ethods of Korenman and D uke12 and Johnson and Nakamura11 w ere used to assay e stro g e n and p ro g e ste ro n e receptor, respectively. Tritiated estradiol or progesterone is allowed to incubate with cytosol at different concentrations o f lig an d. F o llo w in g e q u ilib ra tio n, unbound ligand is rem oved by treatm ent with dextran-coated charcoal. Radioactivity rem aining after the removal of the dextran-coated charcoal is due to receptor bound estrad io l or p ro g estero n e. N on-specific b in d in g is th at b in d in g measurable in the presence of a saturating co n cen tratio n of non-radioactive ligand and is subtracted from the total. The data are analyzed by Scatchard analysis to arrive at num ber and affinity of binding sites. Sixty percent of the tests perform ed in our laboratory are obtained from patients

2 6 8 SZ A K A C S, A R R O Y O A N D G IR G E N T I TABLE I Breast Tumors from St. Joseph's Hospital and Other Tampa Bay Hospitals TABLE II Correlation of Histologic Grade with Estrogen Receptor and Progesterone Receptor Status C ases ER+ ER- P e r c e n t o f C a se s P e r c e n t ER+ P e r c e n t PR+ St. Joseph's 540 435 80.3% 105 19.7% Other 504 387 76.7% 117 23.2% PR+ PR- St. Joseph's 326 191 58.5% 135 41.5% Other 382 202 52.8% 180 47.2% undergoing surgery in our institution. After removal, the tissue is rushed from the operating room to the pathology laboratory w here the tum or is dissected from the rest of the specimen. A representative sample is subm itted for frozen section diagnosis and, w hen possible, one gram of tum or is diced, frozen in liqu id n itro g e n, and sto re d at 72 C. Forty percent of the samples examined consisted of already frozen specim ens. Freezing was achieved in different ways, including the use of a cryostat, freon, (such as Cryokwik), and liquid nitrogen. Based on our own experience and the generally accepted values rep o rte d in the literature, levels in excess of three fmoles per mg of cytosolic protein were considered positive for estrogen receptor. P rogesterone recep to r values are considered positive at values greater than 10 fmoles per mg of cytosolic protein. In those cases where histologic material was available for review, the grading of each tum or was made in accordance with the guidelines suggested by Fisher, Redman and Fisher,5 taking into conside ra tio n th e p re s e n c e o r a b se n c e of tubules in the invasive com ponent of each tum or and the degree of cellular differentiation w ith grade 1 being the most differentiated and grade 3 the least. Only infiltrating ductal carcinomas were selected for correlation w ith recepto r assay v alu es. T he tu m o rs w ere also divided into three categories according Grade 1 13 98 70 Grade 2 28 86 63 Grade 3 59 62 46 to the presence of slight, m oderate, or marked necrosis. Results T he p e rc e n t of p o sitiv e h o rm o n e receptors in our cases w ere com pared to those cases subm itted to us from other hospitals. Estrogen receptor content was m easured in 1,044 cases of which 540 w ere from St. Joseph s H ospital (SJH) and 504 from referring hospitals (RH) (table I). Eighty p ercen t of SJH cases and 77 p e rcen t of th e RH specim ens w ere positive. P rogesterone receptors were m easured in 708 cases. Fifty-eight percent of SJH and 53 percent of the RH cases w ere positive for p ro g estero n e receptors. Com bining both groups, 80.5 p e rc e n t w e re p o sitiv e for e stro g e n receptors, and 56 percent w ere positive for progesterone receptors. The percentage of tum or cells in each sample was estimated in all cases where TABLE III Correlation of Estrogen Receptor+ and Progesterone Receptor+ Status with Histology ER+ PR+ Infiltrating duct, nos 79% (275) 57% (117) Comedo carcinoma 72% ( 26) 29% ( 5) Medullary 70% ( 28) 41% ( 10) Mucinous 90% ( 18) 61% ( 8) Tubular 100% ( 4) 100% ( i) Lobular 91% ( 40) 71% ( 22) Mixed 97% ( 30) 88% ( 23) Papillary 70% ( 7) 50% ( 3) Undifferentiated 62% ( 7) 50% ( 2)

T A B L E IV Tumor Necrosis and Estrogen and Progerterone Receptor Status N e c r o s is Percent ER+ P e r c e n t PR+ A S S E S S M E N T O F E R A N D P R IN B R E A S T C A N C E R 2 6 9 like necrosis seen in the invasive area of the tumor. Tumor positivity for estrogen and p ro g estero n e receptors was in d i rectly p ro p o rtio n al to th e d eg re e of tumor necrosis (table IV). Absent 87 64 Moderate 65 33 Severe 45 22 histologic material was available for evaluation. Except in the very hypocellular specimen, there was no clear correlation betw een the cellularity of the tum or and the positivity of the sample. G rade I, well differentiated carcinoma, was more likely to have estrogen and progesterone receptors than the less differentiated, more aggressive tum ors of Grades II and III (table II). In our study, a greater percentage of mucinous, tubular, lobular, and mixed tumors with lobular elem ents contained greater than three fmoles per mg of protein of estrogen receptors than the other types. A similar trend was noted for prog estero n e recep to rs (table III). This observation has also b e e n m ade and reported in the past by several investigators.2'6,15'17 The presence of necrosis in the histologic material has consistently been associated with poor prognosis.5 Included in the sam e group are th e two types of tum or necrosis found in breast cancer, i.e., the necrosis present in the comedocarcinoma and the less frequent infarct- Survival Statistics Three hundred and twenty-two cases of carcinoma of the breast diagnosed and treated at St. Joseph s Hospital and followed by the Tumor Registry have been analyzed for su rv iv al sta tistic s. T he expected survival rates were taken from th e Table of Survival P robabilities for w om en in th e U.S. published by the National C enter for Health Statistics and based on the mean patient age for each group. The observed survival rates w ere then expressed relative to the published Survival tables.1 Tumor estrogen receptor and progesterone receptor studies were perform ed TABLE V Evaluation of Estrogen and Progesterone Receptor Status Survival Data ER Status Number of Cases Percent of Total Age Range Average Age Percent of Group Over 50 Years ER-, PR- 86 26.6 27-84 55.4 33.7 ER+, PR+ 152 47.4 29-89 63.5 17.0 ER+, PR- 64 19.8 38-88 63.7 12.5 ER-, PR+ 20 6.2 25-82 53.5 45.0 Total 322 100 25-29 60.8 22.3 F i g u r e 2. R e la tio n sh ip b e tw e e n age and ER PR status. M = average age.

270 SZAKACS, ARROYO AND GIRGENTI =»140 130.. -::-:...... ::...::-.. --:: lr-... :.' '20 110 FIGURE 3. Distribution of ER concentration in each decade. ClOD! 2... ~ o 90 > U 80 '"! "0 70 os ~ 60 50.... 40 30 20 10 20 30..... o\\\\i"mu. 60 Age in Decades....... 70. 80,.'40 '30....... ~ ~~.. ar!rffi~:... c.. '00... 2 ;; g,. " u.. 80!;; os Ii '20 110 90 70... 60.,..,., FIGURE 4. Distribution of PR concentration in each decade. 50 40 30 20 '0 0 20 30 40 Age in Decades.., 70 80

T A B L E VI Increase in Estrogen and Progesterone Positivity A ge R ange Percent ER+ P e r c e n t PR+ A S S E S S M E N T O F E R A N D P R IN B R E A S T C A N C E R 271 30-40 31.8 36.4 40-50 54.4 52.2 50-60 62.7 46.7 60-70 68.2 50.0 70-80 88.9 66.6 on all patients. Appropriate therapy and follow-up were provided at SJH. Follow -up studies of these patients ranged from one to 48 m onths from the tim e of diagnosis with 56 patients follow ed for four y ears. T he actu arial m ethod was used to calculate the survival data on a monthly basis. The cause of d eath was not specified. T he age ranged from 25 to 89 years, all patients were female, and there was no distinction made for race. Four groups w ere evaluated based on estrogen and progesterone receptor status as seen on the following table (table V). O f th o se p a tie n ts w ho w ere b o th estro g en re c e p to r and p ro g e ste ro n e receptor negative, one third w ere under the age of 50 with the average age of the group being 55.4 years. The estrogen receptor and progesterone receptor positive patients had an average age of 63.5 years with only 17 percent under the age of 50 years. There is no significant differ- M o nths Follow ed F i g u r e 5. O bserved survival rates of patients with different ER PR status. 0 = E R - P R - = E R + P R + * = ER + P R - ence in age distrib u tio n b etw een the ER + P R + and E R + P R groups, sugg estin g E R p o sitiv ity m ay b e m ore closely related to increasing age than PR status. The group of E R P R + patients is too small to draw any conclusive data, although most of these patients fell into a younger group (figure 2). The concentration of ER increased with age, and those patients over 50 w ere m ore likely to have a strongly positive result (figure 3). OD*D * ER-PR S t a t u s T A B L E V I I Survival Data Based on Estrogen and Progesterone Receptor Status O b s e rv e d S u r v i v a l R a te ( P e r c e n t) E x p e c te d S u r v i v a i R a te ( P e r c e n t) Adjusted S u r v i v a l R a te ( P e r c e n t ) S ta n d a r d E r r o r I3ÜÎ O ER-, PR- 82.9 96.3 86.0 0.053 ER+, PR+ 93.3 92.0 100.0 0.012 ER+, PR- 83.3 92.0 90.5 0.040 ER-, PR+ three year survival rate is 100 percent, the four year data are not significant. Standard error = 0.071. F i g u r e 6. M onths Follow ed A d j u s t e d s u r v i v a l r a t e s o f p a t i e n t s w i t h d i f f e r e n t ER PR s ta tu s. 0 = E R - P R - = ER + PR 4- * = ER + P R -

2 7 2 SZ A K A C S, A R R O Y O A N D G IR G E N T I Concentration of PR increased less dramatically with age (figure 4). There was a significant increase in PR positivity over 40 years of age with no fu rth er increase u ntil 70 years of age while ER positivity increased linearly with age (table VI). The four year survival data are displayed here based on the status of estrogen and progesterone receptors (table VII). For the group E R P R +, only five cases w ere followed beyond three years, and survival statistics could not be calculated on this number. O ur data indicate that patients diagnosed and treated at St. Joseph s Hospital with E R + P R + status have a 93.5 percent observed four year survival rate and 100 percent ± 2. 4 percent relative survival rate (at 95 percent confidence), which is above the national average for this group. Those patients who w ere ER P R o re R + P R had an average observed survival rate of 83 p ercent; how ever, w h en c o rre c te d for o th e r causes of death and age, the relative survival rates diverged. However, at the 95 percent confidence level, this difference is insignificant. The ER PR patients have a relative survival rate of 86.0 perc e n t ± 10 p e r c e n t a n d E R + P R patients have a relative survival rate of 90.5 percent ± 8 percent. These findings suggest that PR status predicts more accurately survival prognosis than ER status (figures 5 and 6). F u rth e r inform ation g ath ered from the survival studies reveals a significant drop in survival b etw een 12 and 30 months for all groups. This suggests a period of increased risk for recurrence in all types of carcinoma. The mortality rate observ ed for th e e n tire study of 322 cases over four years was 9.31 percent. The mortality rates by group were 15.1 percen t for th e ER P R group, 4.6 percent for the E R + P R + group, 14.1 percent for the E R + P R group, and 5.0 percent for the ER P R + group. All deaths occurred prior to 30 m onths of follow-up. Conclusion The percentage of tum or epithelium present in the sample was not proportional to th e levels of recep to rs. O ur total rate of positive estrogen and progesterone receptors, as well as the proportions of the different tum or phenotypes, are in accord w ith m ost of the values published in the m edical literatu re.17,19 As in most series, our older, post-m enopausal patients had tum ors that were more frequently steroid receptor positive.16 P ro g estero n e recep to rs have b een m entioned previously as the m ore specific of the two m arkers,4 and our data seem to support this theory. Efforts are made to report levels of both tumor steroid receptors. Based on our review of 322 cases of breast carcinoma at St. Joseph s Hospital from January 1979 to D ecem ber 1983, there is a significant correlation betw een survival and PR status. The prognostic value of the PR assay as a single test is superior to the ER assay, although ER status alone correlates m ore closely to age. The group ER P R + is small in num ber and thus unsuitable for analysis; however, others have suggested that this group may represent an error of the ER test itself which is not as sensitive as the PR assay. These patients, although few, did have very high survival rates and perhaps should be considered to be in the E R + P R + category for prognostic purposes.18,19 The high survival of the ER + P R + patients is very encouraging for the recent advances in antiestrogen therapies. F urther follow-up may reveal substantial cure rates for this group. Acknowledgments Thanks are extended to Mrs. Cheryl Colhouer for her clerical and editorial assistance in the preparation

of the manuscript. In addition, thanks are extended to Mr. Warren Chandler for his assistance in processing the photographic material. References 1. American Joint Committee for Cancer Staging and End-Result Reporting, Manual for Staging of Cancer, 2nd ed., 1983, pp. 17 19. 2. A n t o n i a d e s, K. and SPECTOR, H.: Correlation of estrogen receptor levels with histology and cytomorphology in human mammary cancer. Amer. J. Clin. Path. 71:497-503, 1979. 3. B e a t s o n, G. T.: On the treatm ent of unoperative cases of carcinoma of the mamma. Suggestions for a new method with illustrative cases. Lancet. 2:104-107, 1986. 4. C l a r k, G. M., et al: Progesterone receptors as a prognostic factor in Stage II breast cancer. New Engl. J. Med. 309:1343-1347, 1983. 5. F i s h e r, E. R., R e d m o n d, C., and F i s h e r, B.: Pathologic findings from the national surgical adjuvant breast project (Protocol #4). VI. Discrim inants for five-year tre a tm e n t failure. Cancer 46:908-918, 1980. 6. Howat, J. M. T., et al: The association of cytosol estrogen and progesterone receptors with histological features of breast cancer and early recurrence of disease. Brit. J. Cancer 47:629 640, 1983. 7. H u b a y, C. A., et al: A ntiestrogen, cytotoxic chem otherapy and bacillus Calm ette-g uerin vaccination in Stage II breast cancer. A preliminary report. Surgery 87:494-501, 1980. 8. HUBAY, C. A., et al: Hormone receptors, an update and application. Symposium on breast cancer. Surg. Clin. North Am. 64:1155 1171, 1984. A S S E S S M E N T O F E R A N D P R IN B R E A S T C A N C E R 2 7 3 9. J e n s e n, E. V., et al: E strogen recepto r and breast cancer response in adrenalectomy. Prediction in Response in Cancer Therapy. Hall, T. C., ed. M onograph 34. B eth esd a, M D, National Cancer Institute, 1 9 7 1. 10. J ia n g, N.S.: Breast cancer: Estrogen and progesterone receptor assays as a guide to therapy. Mayo Clin. Proc. 58:64, 1983. 11. J o h n s o n, R. B. and N a k a m u r a, R. M.: Simplified Scatchard plot assay for progesterone receptor in breast cancer: Comparison with singlepoint and m ulti-po int assay. C lin. C hem. 24:1170 1176, 1978. 12. K o r e n m a n, S. G. and D u k e s, B. A.: Specific estrogen binding by the cytoplasm of human breast cancer. J. Clin. Endocrinol. 30:639-643, 1970. 13. M a n n i, A.: H orm one recepto rs and breast cancer. New Engl. J. Med. 309:1383-1385, 1983. 14. O X LEY, D. K.: H orm one receptors in breast cancer analytic accuracy of contemporary assays. Arch. Pathol. Lab. Med. 108:20-23, 1984. 15. P a r l, F. L. and W a g n e r, R. K.: The histopathological evaluation of human breast cancers in co rrelation w ith estrogen re c e p to r values. Cancer 46:362-367, 1980. 16. Q a z i, R., et al: Estrogen receptors and the pattern of relapse in breast cancer. Arch. Int. Med. 144:2365-2367, 1984. 17. R o s e n, P. P., et al: Pathological review of breast lesions analyzed for estrogen receptor protein. Cancer Res. 35:3187-3194, 1975. 18. S a r r i f, A. M. and D u r a n t, J. R.: Evidence that estrogen receptor negative, progesteron receptor positive breast and ovarian carcinomas contain estrogen receptor. Cancer 48:1215-1220, 1981. 19. W lt T L lf F, J. L.: Steroid-hormone receptors in breast cancer. Cancer 53:630-643, 1984.