`
`Sequential steroid hormone receptor
`measurements in primary breast cancer
`with and without intervening primary
`chemotherapy
`
`S Taucher, M Rudas‘, M Gnant, K Thomanek, P Dubsky, S Roka,
`T Bachleitner, D Kandioler, C Wenzel2, G Steger2, M Mittlbéck
`and R Jakesz
`
`Department of Surgery, ‘Institute of Pathology and 2First Medical Department, Vienna University Medical
`School, Waehringer Guertel 18-20, Vienna A—1090, Austria
`(Requests for offprints should be addressed to S Taucher,‘ Email: susanne.taucher@ univie.ac.at)
`
`Abstract
`
`The objective of this analysis was to determine the accuracy of steroid receptor measurement in large
`core needle biopsies compared with surgically removed specimens and the influence of preoperative
`chemotherapy on hormone receptor status. We consecutively performed 722 large core needle
`biopsies in palpable lesions of the breast. The diagnosis of breast cancer was confirmed upon biopsy
`in 450 patients; 236 women underwent immediate surgery, and 214 patients received preoperative
`chemotherapy. We assessed estrogen (ER) and progesterone receptor (PR) in biopsy tissue and
`surgically removed specimens and calculated accuracy, sensitivity, specificity, the weighted K value
`and Spearman’s rank correlation. The modulation of steroid receptor status in preoperatively treated
`patients was tested by Cochran—Mante|—Haensze| statistics. The accuracy of ER evaluation in the
`biopsy material of patients without intervening chemotherapy was 91%, sensitivity and specificity
`were 94% and 80% respectively. Accuracy, sensitivity and specificity were 86% in patients treated
`preoperatively. In terms of PR assessment, we obtained slightly inferior results: accuracy, sensitivity
`and specificity were 80%, 73% and 85% respectively in patients without preoperative treatment, and
`79%, 48% and 92% respectively in patients undergoing preoperative therapy. Following preoperative
`chemotherapy, patients showed a significant increase in ER—negative (P=0.02) and PR—negative
`(P: 0.0005) measurements. We have concluded from our
`results that ER and PR receptor
`measurement in core needle biopsy is a reliable basis in clinical practice for selecting patients for
`neoadjuvant endocrine treatment. Preoperative cytotoxic chemotherapy induced a significant extent
`of variation in the steroid receptor expression of breast cancer cells.
`Endocrine-Re/ated Cancer (2003) 10 91-98
`
`lntroductlon
`(PR) determi-
`Estrogen (ER) and progesterone receptor
`nations are established procedures in the routine management
`of patients with breast cancer, chiefly as predictive factors
`for response to adjuvant and palliative endocrine therapy
`(Clark 1996, Allred el al. 1998, Harvey el al. 1999). In
`addition, the selection for preoperative treatment modalities
`needs to be directed by molecular markers such as steroid
`hormone receptor status. Mouridsen el al. (1978), Allegra el
`al. (1980) and Chang el al. (1999) have shown that pretreat-
`ment ER and PR values significantly predict the response to
`preoperative administration of tamoxifen.
`
`The selection for preoperative chemotherapy is currently
`based on such clinical factors as tumor size more so than on
`molecular markers. Preoperative chemotherapy has shown to
`increase the percentage of breast conservation, but does not
`positively correlate with overall survival (Fisher el al. 1998,
`Wolmark el al. 2001). In addition to the absence of c—erbB—2,
`lack of ER has recently been demonstrated to significantly
`predict for subsequent good clinical response. Lack of ER
`expression was additionally predictive for increased risk of
`death (Chang el al. 1999).
`Studies from our own group in predictive factors indicat-
`ing response to primary chemotherapy have shown that Her2/
`neu overexpression is predictive for achieving a pathological
`
`End0crine—ReIated Cancer (2003) 10 91-98
`1351-0088/03/010-091 © 2003 Society for Endocrinology Printed in Great Britain
`
`Online version via http://www_endocrino|ogy_org
`
`|nnoPharma Exhibit 1073.0001
`
`
`
`Taucher et al..' Hormone receptor status in breast cancer
`
`complete response after a preoperative taxane—containing
`cytotoxic chemotherapy regimen (Steger el al. 2000).
`Together with clinical characteristics, assessment of bio-
`logical and molecular markers pretherapy should allow phys-
`icians to focus therapeutic considerations on a given patient’s
`individual tumor factors. Quality assurance of these measure-
`ments is the main framework for these considerations. A
`
`nationwide external quality assurance project has been con-
`ducted to assess the quality of immunohistochemical steroid
`receptor evaluation (Regitnig el al. 2002). The results of this
`quality program showed an excellent concordance of hor-
`mone receptor assessments in terms of ER (K=0.57) and
`slightly inferior results with regard to PR (K: 0.53).
`The primary objective of this analysis was to determine
`the accuracy of steroid receptor measurement in large core
`needle biopsies (CNB) compared with surgically removed
`specimens. Large core biopsy is a valid tool for the preoper-
`ative management of breast lesions (Di Loreto el al. 1996).
`The concordance of ER and PR status in biopsy and final
`specimen in a preoperatively untreated group allows for per-
`fect quality assurance to assess ER and PR in CNB. The
`response of the primary tumor and lymph node metastasis to
`preoperative cytotoxic treatment
`is
`the most
`important
`parameter for overall outcome. Therefore, knowledge of fac-
`tors predicting tumor response can avoid administration of
`treatment to patients who are not likely to respond. The
`second aim of this analysis was to determine the influence
`of preoperative chemotherapy on steroid receptor status. The
`changes in biological markers induced by chemotherapy may
`lead to a better understanding of breast cancer biology.
`
`Materials and methods
`
`Between 1994 and 2000, we evaluated 722 patients with a
`palpable mass in the breast in an attempt to determine the
`histology of the lesion and — in the case of breast cancer —
`to compare several prognostic markers in tumor
`tissue
`obtained preoperatively from CNB and surgical specimens
`within a prospective evaluation. Pretherapeutic mammo-
`graphy was present and CNB was performed under local
`anesthesia using a 15—gauge needle (ASAP Detachable
`Biopsy System, Boston Scientific Corporation, Vienna,
`Austria) in all patients.
`Of these patients, 450 (62.3%) presented with an epithelial
`malignancy of the breast upon biopsy. No false—positive
`results were seen. Immediate surgery was administered to
`236 patients (52.4%) and sufficient material to compare hor-
`mone receptor status in the biopsy tissue and in the surgically
`removed specimen was obtained in 180 patients (76.3%). In
`order to facilitate breast—conserving surgery, 214 patients
`(47.6%)
`received preoperative chemotherapy. Excellent
`responses to primary treatment and complete pathological
`remission were observed in 23 women (10.7%), 191 patients
`were thus remaining in this analysis of sequential hormone
`
`receptor assessment in biopsy and surgically removed tissue.
`The two patient groups were analyzed separately.
`CNB specimens were transferred to 24—h fixation in neu-
`tral—buffered formalin, and paraffin—embedded tissue sections
`of 3 pm thickness were processed. ER and PR determinations
`were performed by immunohistochemistry as described ear-
`lier (Reiner el al. 1990). In brief, tissue sections stained with
`a cut—off of less than 10% and low intensity of staining were
`recorded as hormone receptor negative, those showing more
`than 10% stained tissue were considered as receptor positive,
`showing three qualities discriminated as follows: weakly
`positive (1 +), 10-50% showed stained tumor tissue (medium
`positive (2 +), 50-80%; strongly positive (3 +)) and those
`with more than 80% staining. Intensity of staining was also
`recorded, influencing the results according to a distinct scor-
`ing scale (Reiner score) (Reiner el al. 1990). All
`tissue
`samples were scored by a single pathologist (M R), blinded
`to the intervention with chemotherapy. The methods of the
`whole procedure were not changed over the observation
`period.
`Patients suitable for breast conservation without metast-
`
`ases shown by X—ray, ultrasound and scintigraphy were
`treated with primary surgery. Patients with T3 or T4 lesions,
`those primarily not suited for breast conservation, and those
`presenting with clinical signs of lymph node metastasis
`received preoperative chemotherapy. All women treated with
`breast—conserving surgery received postoperative irradiation,
`unless the patient was involved in a prospective randomized
`trial evaluating the importance of postoperative irradiation in
`a certain patient selection with very low risk for local relapse.
`In patients undergoing modified radical mastectomy, post-
`operative radiotherapy was left
`to the discretion of the
`responsible physician based on consultation within the
`interdisciplinary team.
`The characteristics of the patients without preoperative
`treatment are shown in Table 1. The median age was 62
`years, patients were predominantly postmenopausal (81 .7%),
`40% had a tumor smaller than 2 cm in diameter, and 47.2%
`had no lymph node involvement. Breast conservation was the
`predominant surgical procedure and was performed in 55%.
`The characteristics of patients receiving preoperative
`chemotherapy are described in Table 2. Fluorouracil, epirub—
`icin and cyclophosphamide (CMF) therapy was used primar-
`ily as it was part of a clinical study, the Austrian Breast and
`Colorectal Cancer Study Group Trial 7 (Iakesz 2001). An
`anthracycline—containing regimen CMF was subsequently
`introduced for preoperative treatment. Several patients
`(41.8%) received the combination of anthracycline and taxo-
`tere. Response to chemotherapy was assessed according to
`the International Union Against Cancer guidelines (Hayward
`el al. 1977). Complete pathological remission was defined as
`complete disappearance of invasive tumor cells, irrespective
`of a possibly residual, yet exclusively intraductal component.
`Reduction of tumor size of at least 50% was defined as
`
`92
`
`www_endocrino|ogy_org
`
`|nnoPharma Exhibit 10730002
`
`
`
`Table 1 Patient characteristics (without preoperative
`treatment) (n= 180)
`
`Table 2 Patient characteristics (with preoperative treatment)
`(n = 191)
`
`Endocrine-Related Cancer (2003) 10 91-98
`
`Numbers
`
`Percentage
`
`Characteristics
`
`Numbers
`
`Percentage
`
`Characteristics
`
`Age (years)
`Median
`
`Range
`< 50
`> 50
`Clinical tumor size (cm)
`< 2
`2-5
`> 5
`
`62
`
`23-88
`33
`147
`
`72
`98
`10
`
`18.3
`81.7
`
`40.0
`54.4
`5.6
`
`Age (years)
`Median
`
`Range
`Premenopausal
`Postmenopausal
`Clinical tumor size
`T1
`T2
`T3
`
`T4
`Preoperative therapy
`CMF
`FEC
`Taxane-containing regimen
`Response to chemotherapy
`pPR
`pNC
`pPD
`Surgical procedure
`Breast conservation
`Mastectomy
`Pathological tumor stage
`
`pT3
`pT4
`
`51
`
`33-74
`73
`118
`
`5
`110
`42
`
`34
`
`58
`59
`74
`
`113
`74
`4
`
`126
`65
`
`20
`18
`
`38.2
`61.8
`
`2.6
`57.6
`22
`
`17.8
`
`30.4
`30.9
`38.7
`
`59.2
`38.7
`2.1
`
`66
`34
`
`105
`9.4
`
`99
`81
`
`72
`73
`12
`15
`8
`
`55.0
`45.0
`
`40.0
`40.6
`6.7
`8.3
`4.4
`
`Surgical procedure
`Breast conservation
`Mastectomy
`Pathological tumor stage
`pT1
`pT2
`pT3
`pT4
`pTX
`Pathological nodal stage
`pN0
`85
`47.2
`pN1
`77
`42.8
`pNX
`18
`10.0
`pT1—X, pNQ—X, according to UICC criteria (International
`Union against Cancer).
`
`partial remission, reduction of tumor size of less than 50%
`was considered as stable disease, any increase in tumor size
`.
`.
`.
`in the course of preoperative therapy was determined as pro—
`gressive disease.
`Within 7 years, a total of 214 patients was treated preop-
`eratively at our institution. It is noteworthy that these patients
`were generally 10 years younger (mean=51.3 years) than
`those not given primary chemotherapy, only 60% in this
`group were postmenopausal. A tumor of less than 2 cm in
`diameter was present only in 2.8%, and 17% had a T4 lesion.
`Sixty—two patients were treated with CMF as a part of the
`above—mentioned Austrian Trial 7 (Jakesz 2001). Since 1996,
`when increasing evidence was presented that an anthracyc—
`line—containing chemotherapeutic regimen showed higher
`response rates,
`this kind of chemotherapy was mainly
`employed. Eighty—eight patients were part of two clinical
`trials, one with preoperative Taxol and the other with a com-
`bination of epirubicin and taxotere. The overall response to
`primary chemotherapy was 63.4%, the pathological complete
`response rate was 10.7%. Breast—conserving surgery was
`given to 67.6% of all patients given primary chemotherapy.
`
`Statistical analysis
`
`We calculated the accuracy, sensitivity, specificity and K
`coefficient of ER and PR results and correlated positive
`tumors (1 +, 2 + and 3 +) versus negative (0). Accuracy refers
`to the degree of concordance between the percentage of ER—
`
`Pathological nodal Stage
`398
`76
`PNO
`58.1
`1 11
`pN1
`2 1
`4
`pNX
`FEC, fluorouracil, epirubicin and cyclophosphamide; pPR,
`pathological partial remission; pNC, pathologically no change;
`pPD, pathologically progressive disease.
`
`or PR—positive and —negative results in biopsy and the ER or
`PR results in the final histology of surgically removed tissue.
`Test sensitivity was calculated as the percentage of ER— and
`PR—positive biopsies
`in surgically removed specimens
`(percentage true positive). Test specificity was calculated as
`the percentage of ER— and PR—negative biopsies in surgically
`removed tissue (percentage true negative). Patients with or
`without preoperative treatment were analyzed separately.
`K is the proportion of agreements after chance agreement
`has been excluded. Its upper limit is +1.00 (total agreement).
`The value of K is near to zero if agreement between two
`different variables is just by chance. Weighted K was used to
`dichotomize between different categories of steroid receptor
`(0, 1+, 2+, 3 +) (Cohen 1960).
`Spearman’s rank correlation is a distribution—free analog
`of Pearson’s correlation coefficient. Spearman’s rho co-
`efficient (r) indicates agreement. A value of r approximating
`one indicates good agreement; a value near zero poor agree-
`ment. We calculated the correlation of ER and PR results
`
`www_endocrino|ogy_org
`
`93
`
`|nnoPharma Exhibit 10730003
`
`
`
`Taucher et al..' Hormone receptor status in breast cancer
`
`discriminating four different qualities, negative, 1 +, 2 + and
`3 + positive.
`The influence of primary chemotherapy on steroid recep-
`tor status in the final specimen was tested by Cochran-
`Mantel—Haenszel statistics. This statistic assumes a common
`
`the purpose of the Cochran—Mantel—
`in short,
`odds ratio;
`Haenszel statistic is to test whether the response is con-
`ditionally independent of the hormone receptor status when
`adjusting for the hormone status at biopsy (Landis el al.
`1978).
`
`and 75 (41.6%) in the surgical specimen. Concordant PR
`status in CNB and the final specimen was found in 143 out
`of 180 patients, with an accuracy of 80%. The PR status upon
`biopsy was true positive in 60 patients and true negative in
`83 patients, sensitivity and specificity were 73% and 85%
`respectively. The weighted K for four PR receptor qualities
`was 0.52. The correlation of sequential PR status was also
`significant, although slightly inferior compared with the ER
`results, r = 0.64 (P = 0.01).
`
`Resufls
`
`Patients without intervening preoperative
`therapy
`Results of ER assessment are shown in Table 3. ER was
`
`negative in the biopsy in 50 (27.8%) out of 180 patients
`given neither chemotherapy nor any other cytotoxic treat-
`ment preoperatively, and negative in surgical specimen in 47
`patients (26.1%); 123 patients were regarded to be positive
`in both the biopsy and surgical procedure. Concordant results
`between ER in the biopsy and surgically removed specimens
`were obtained in 163 out of 180 patients, giving an accuracy
`of 91%. Sensitivity and specificity of ER assessment in CNB
`were 94% and 80% respectively. The weighted K was 0.69.
`A significant Spearman’s correlation of sequential ER status
`was found with r: 0.76 (P : 0.01).
`
`Table 3 ER in patients without intervening chemotherapy
`
`ER — biopsy
`
`ER — surgical specimen
`
`Total
`number
`
`Negative 1 +
`
`2 +
`
`3 +
`
`Negative
`1 --
`2——
`3 --
`
`Total number
`
`40
`4
`3
`0
`
`47
`
`5
`14
`4
`5
`
`28
`
`3
`6
`53
`7
`
`69
`
`2
`1
`10
`23
`
`36
`
`50
`25
`70
`35
`
`180
`
`Results of PR assessment are given in Table 4. In terms
`of PR, 98 patients (54.4%) showed negative results in the
`primary biopsy and 105 patients (58.3%) in the surgical
`specimen. Eighty—two (45.6%) were positive in the biopsy
`
`Patients with preoperative treatment
`
`Table 5 presents the results of ER assessment. Out of 191
`patients given intervening chemotherapy, 91 (47.6%) were
`ER negative in the primary biopsy and 92 (48.2%) in the final
`surgical specimen. One hundred patients were ER positive in
`the biopsy and 99 patients in the final surgical material.
`Exactly concordant ER status was found in 164 patients,
`accuracy was 86%. Sensitivity and specificity of ER evalu-
`ation was 86% and 86% respectively. The weighted K was
`0.64. Spearman’s coefficient was significant, r=0.75 (P:
`0.01).
`
`Table 5 ER in patients with preoperative chemotherapy
`
`ER — biopsy
`
`ER — surgical specimen
`
`Total
`number
`
`Negative 1 +
`
`2 +
`
`3 +
`
`Negative
`1 --
`2 --
`3 --
`
`Total number
`
`78
`9
`5
`0
`
`92
`
`7
`7
`8
`3
`
`25
`
`2
`8
`30
`3
`
`43
`
`4
`2
`12
`13
`
`31
`
`91
`26
`55
`191
`
`191
`
`Results of PR assessment are given in Table 6. One hun-
`dred and thirty—one patients (68.6%) showed PR—negative
`results in the biopsy and 152 women (79.6%) were finally
`identified as PR negative. PR was shown to be positive in
`both biopsy and surgically removed material
`in only 29
`patients. The accuracy of PR measurement in needle biopsy
`was 79%. Only 29 out of 60 patients showed true—positive
`PR status in the biopsy, and sensitivity was rather low at
`48%. The number of true—negative biopsies was 121 out of
`131 with a specificity of 92%. Sequential PR assessment after
`
`Table 4 PR in patients without intervening chemotherapy
`
`Table 6 PR in patients with preoperative chemotherapy
`
`PR — biopsy
`
`PR — surgical specimen
`
`Total
`number
`
`PR — biopsy
`
`PR — surgical specimen
`
`Total
`number
`
`Negative 1 +
`
`2 +
`
`3 +
`
`Negative 1 +
`
`2 +
`
`3 +
`
`Negative
`1 ——
`2 ——
`3 ——
`
`83
`14
`7
`1
`
`Total number
`
`105
`
`8
`12
`9
`2
`
`31
`
`7
`6
`16
`5
`
`34
`
`0
`1
`5
`4
`
`98
`33
`37
`12
`
`Negative
`1 ——
`2 ——
`3 ——
`
`10
`
`180
`
`Total number
`
`121
`18
`9
`4
`
`152
`
`3
`8
`0
`4
`
`4
`4
`7
`2
`
`15
`
`17
`
`3
`1
`2
`1
`
`7
`
`131
`31
`18
`1 1
`
`191
`
`94
`
`www_endocrino|ogy_org
`
`|nnoPharma Exhibit 1073.0004
`
`
`
`Endocrine-Related Cancer (2003) 10 91-98
`
`preoperative chemotherapy produced a rather low level of
`correlation, weighted K was 0.37. However, the calculation
`of Spearman’s coefficient did indicate a significant correla-
`tion, r = 0.47 (P = 0.01).
`
`tested again by the Cochran—Mantel—Haenszel statistics.
`Menopausal status offers no explanation of the decrease in
`PR status, only eight (26%) of the primarily PR-positive
`patients were premenopausal at the time of diagnosis and
`developed PR—negative disease following chemotherapy.
`
`Modulation of ER status in patients with
`ER—positive results in CNB (Table 7)
`
`Discussion
`
`Fourteen out of 100 ER—positive patients undergoing pre-
`operative treatment showed a decrease in ER status and
`finally proved to be ER negative (14%), whereas only seven
`out of 130 ER—positive patients without preoperative
`chemotherapy were ER negative in the surgically removed
`specimen (5.4%). The calculation of Cochran—Mantel—
`Haenszel statistics demonstrated a statistically significant
`shift of ER—positive to ER—negative status due to primary
`chemotherapy (P = 0.02).
`
`Table 7 Modulation of ER status in patients with ER—positive
`results in CNB
`
`ER status in final specimen
`Preoperative
`chemotherapy %. .
`.
`ER negative
`ER positive
`
`Total ER
`positive
`in CNB
`
`Without
`With
`
`Total
`
`7 ( 5.4%)
`14 (14.0%)
`
`123 (94.6%)
`86 (86.0%)
`
`21 ( 9.1%)
`
`209 (90.9%)
`
`130
`100
`
`230
`
`The endocrine effect of neoadjuvant chemotherapy
`inducing menopause in premenopausal patients may be a
`simple explanation of decreasing ER expression in these
`patients. In fact, seven out of fourteen (50%) patients show-
`ing a decrease of ER status were premenopausal at the time
`of diagnosis.
`
`Modulation of PR status in patients with
`PR-positive results in CNB (Table 8)
`
`Fifty—eight patients were primarily PR positive in the core
`biopsy, 30 out of 58 (51.7%) presented with PR negativity
`in the final surgical specimen following preoperative treat-
`ment. In patients not given preoperative therapy, only 22 out
`of 82 PR-positive women (26.8%) finally appeared to be PR
`negative. The increase in PR negativity induced by preoper-
`ative chemotherapy was statistically significant (P= 0.0005)
`
`Table 8 Modulation of PR status in primarily PR-positive
`patients
`
`Preoperative
`chemotherapy
`
`PR status in final specimen
`.
`.
`.
`PR negative
`PR positive
`
`Total PR
`positive
`in CNB
`
`Without
`With
`
`Total
`
`22 (26.8%)
`30 (51.7%)
`
`60 (73.2%)
`28 (48.3%)
`
`52 (37.1%)
`
`88 (62.9%)
`
`82
`58
`
`140
`
`The results of sequential steroid hormone receptor assess-
`ment upon immunohistochemistry in CNB and a definite sur-
`gical specimen, without intervening treatment performed at a
`time interval of 7-14 days, failed to show any marked differ-
`ence. Disconcordance between measurements at these differ-
`
`ent time—points was less than 10% for ER and approximately
`20% for PR determinations. In terms of ER negativity in sur-
`gical specimens, correct prediction from biopsy was possible
`in 40 out of 47 (85.1%), and for positivity in 123 out of 133
`patients (92.5%). As to PR negativity, the correct prediction
`from CNB was 83 (79.0%) out of 105 in surgically removed
`material, and in 59 (79.7%) out of 74 patients for PR positiv-
`ity. It can therefore be stated that receptor measurements in
`CNB are, in a high percentage, representative of the receptor
`quality shown by the entire tumor.
`Steroid hormone receptors have proven to be the most
`important predictive markers for selection of systemic treat-
`ment. This has been shown clearly in the selection of postop-
`erative endocrine treatment for premenopausal as well as
`postmenopausal patients (Early Breast Cancer Trialists’ Col-
`laborative Group 1998a, Iakesz el al. 2002). Data from the
`overview concerning the effect of dependence on tamoxifen
`on receptor status indicated a risk reduction in the annual
`odds of recurrence of 34% and 10% for ER—positive and ER-
`poor patients respectively. The reduction in the annual odds
`of death was 20% and only 6% at 10 years for the two differ-
`ent receptor qualities (Early Breast Cancer Trialists’ Collab-
`orative Group 1998a).
`Whether or not receptor status plays an important role as
`a predictive marker for response to chemotherapy is still an
`open question. In metastatic breast cancer, Lippman & Alle-
`gra (1980) have argued that patients with ER—negative tumors
`show a much better response rate than those with ER—positive
`tumors, although this result was not confirmed by other
`authors. In the adjuvant situation, the overview data can be
`interpreted such that chemotherapy appears more beneficial
`in receptor—negative patients (risk ratio of annual odds of
`recurrence was 36% compared with only 20% in ER—positive
`patients), yet the formal test for heterogeneity was not sig-
`nificant in that meta—analysis.
`A retrospective analysis presented at the National Insti-
`tutes of Health (NIH) Consensus Conference in 2000 indicated
`that only patients with ER—negative tumors drew a significant
`benefit from adjuvant taxane administration (NIH 2001). It
`should therefore be strongly recommended that future adjuvant
`trials be based on the quality of steroid hormone receptor
`
`www_endocrino|ogy_org
`
`95
`
`|nnoPharma Exhibit 10730005
`
`
`
`Taucher et al..' Hormone receptor status in breast cancer
`
`status in order to prospectively evaluate their predictive
`importance for response to adjuvant chemotherapy. Knowl-
`edge of receptor status is essential to conduct trials applying
`preoperative endocrine treatment.
`Another goal of our study was to investigate whether or
`not primary chemotherapy induces some selective change in
`the hormone receptor distribution of the primary tumor.
`Breast cancer is a heterogeneous disease in terms of the qual-
`ity of different cell clones present within the same tumor.
`This applies to many molecular and biological properties
`indicated by hormone receptors and other measurable mark-
`ers. As discussed earlier, we hypothesized that primary
`chemotherapy preferentially kills undifferentiated ER— and
`PR—negative tumor cells and largely tends to leave ER—
`positive and PR—positive tumor cells behind. This assumption
`is supported by results from the Early Breast Cancer Trial-
`ists’ Collaborative Group (1998b) investigating recurrence
`and death rates in patients treated with adjuvant chemo-
`therapy and stratified by ER content of the primary tumor.
`The data indicate that the reduction in the annual odds of
`
`recurrence is 30% in ER—poor patients and 18% in ER—
`positive patients respectively. The reduction in the annual
`odds of death in ER—negative patients amounts to 17%, while
`there is no significant improvement in overall survival for
`ER—positive patients.
`If adjuvant chemotherapy were more effective in recep-
`tor—negative tumors, receptor status would change subsequent
`to primary chemotherapy. However, we failed to find a corre-
`lation to support this hypothesis. The number of tumors
`determined as ER negative increased significantly in patients
`undergoing preoperative
`chemotherapy. No
`exhaustive
`explanation is available as yet for these unexpected results.
`The fact that receptor content cannot logically be measured
`in a tumor after a complete pathological response is another
`problem of methodology. The numbers we report, however,
`make it unlikely that the results are significantly influenced
`by this potential source of bias.
`Results similar to our own were reported in very small
`patient cohorts by Frassoldati el al. (1997) and Makris el
`al. (1999). Responders to primary chemotherapy presented a
`significant decrease in ER levels, and all patients showed
`significant increases in apoptotic index and p170 regardless
`of type of response (Frassoldati el al. 1997). Reduction in
`ER scores, but not in PR, was reported for responders to
`neoadjuvant chemotherapy assessed by sequential fine needle
`aspiration (Makris el al. 1999). However, several other
`studies investigating modulation of steroid receptor status by
`primary chemotherapy reported no significant changes of ER
`or PR (Hawkins el al. 1990, Bottini el al. 1996, Schneider el
`al. 2000).
`The majority of studies investigating the biological
`mechanism of steroid receptor expression and regulation
`are initiated to clarify the development of resistance to
`endocrine treatment. Data are currently lacking as to hor-
`
`mone receptor alteration due to cytotoxic substances. Rob-
`ertson (1996) has proposed that estrogen is a stable pheno-
`type in breast cancer cells. While expression of ER in
`tumor cells is stable,
`the relative or absolute number of
`ER—positive or ER—negative cells may vary in the course of
`disease, depending on a variety of host—tumor interactions.
`Epidermal growth factor receptor
`(EGFR) expression is
`inversely related to ER expression,
`subdividing
`ER—
`negative
`tumors
`into two groups; ER—negative/EGFR—
`negative tumors are more likely to respond to endocrine
`treatment
`than
`ER—negative/EGFR—positive
`tumors
`(Nicholson 1993). The dual receptor phenotype may not
`be irreversibly fixed. Mutually exclusive staining for ER
`or EGFR on individual tumor cells raises the option that
`ER and EGFR expression either have a common regulating
`mechanism or that both pathways interact to cross—regulate
`their expressions (Sharma el al. 1994). The major con-
`trolling mechanism of EGFR overexpression in breast
`cancer cells is transcriptional regulation. Wilson & Chryso—
`gelos (2002) have identified a region within the first intron
`of the EGFR gene that mediates transcriptional repression
`of EGFR gene
`expression in ER—positive/EGFR—low—
`expressing breast cancer cells.
`An attractive hypothesis to explain the progression to ste-
`roid independence is that the tumor acquires the ability to
`constitutively express autocrine growth factors. There is evi-
`dence in some cancer models that particular fibroblast growth
`factors (FGF) may function as autocrine growth factors cap-
`able of conferring steroid independence. FGF overexpression
`in the estrogen—dependent breast cancer cell
`line MCF—7
`induces an estrogen—independent phenotype as determined by
`tumor growth and metastasis in nude mice. Because this
`alteration is not due to changes in ER levels, it is likely that
`the FGF autocrine loop acts downstream from an estrogen
`signal (McLeskey el al. 1998).
`Clinical data indicating the potential influence of chemo-
`therapy upon ER status are outstanding as yet. In vilro obser-
`vations using arsenic trioxide (AS203) as a cytotoxic agent,
`however, suggest that AS203 specifically inhibits the expres-
`sion and signaling pathway of EROL (Chen el al. 2002). EROL
`is thought
`to function as a ligand—activated transcription
`factor and promotes growth of breast cancer cells by tar-
`geting expression of signaling components of the insulin—like
`growth factor system (Oesterreich el al. 2001).
`We conclude from our results that the correct prediction
`of ER (91%) and PR (80%) measurement in CNB can indeed
`serve as a reliable basis in clinical practice for selecting
`patients for preoperative systemic treatment. The results are
`less concordant when discriminating the three levels of
`receptor positivity. Quality control for both ER and PR
`measurement is still an important undertaking for the basis of
`daily clinical practice. Surprisingly, primary chemotherapy
`induced a significant decrease in ER and PR levels. Further
`trials are to be designed to clarify the obviously important
`
`96
`
`www_endocrino|ogy_org
`
`|nnoPharma Exhibit 10730006
`
`
`
`interaction between receptor biology and cytotoxic chemo-
`therapy.
`
`References
`
`Allegra JC, Lippman ME, Thompson EB, Simon R, Barlock A,
`Green L, Huff KK, Do HM, Aitken SC & Warren R 1980
`Estrogen receptor status: an important variable in predicting
`response to endocrine therapy in metastatic breast cancer.
`European Journal of Cancer 16 323-331.
`Allred DC, Harvey JM, Berardo M & Clark GM 1998 Prognostic
`and predictive factors in breast cancer by immunohistochemical
`analysis. Modern Pathology 11 155-168.
`Bottini A, Berruti A, Bersiga A, Brunelli A, Brizzi MP, Marco
`BD, Cirillo F, Bolsi G, Bertoli G, Alquati P & Dogliotti L 1996
`Effect of neoadjuvant chemotherapy on Ki67 labelling index,
`c—erbB—2 expression and steroid hormone receptor status in
`human breast tumours. Anticancer Research 16 3105-3110.
`Chang J, Powles TJ, Allred DC, Ashley SE, Clark GM, Makris A,
`Assersohn L, Gregory RK, Osborne CK & Dowsett M 1999
`Biologic markers as predictors of clinical outcome from
`systemic therapy for primary operable breast cancer. Journal of
`Clinical Oncology 17 3058-3063.
`Chen GC, Guan LS, Hu WL & Wang ZY 2002 Functional
`repression of estrogen receptor a by arsenic trioxide in human
`breast cancer cells. Anticancer Research 22 633-638.
`Clark GM 1996 Prognostic and predictive factors. In Diseases of
`the Breast, pp 461-485. Eds JR Harris, ME Lippman, M
`Morrow & S Hellman. Philadelphia, USA: Lippincott—Raven.
`Cohen J 1960 A coefficient of agreement for nominal scales.
`Educational and Psychological Measurement 20 37-46.
`Di Loreto C, Puglisi F, Rimondi G, Zuiani C, Anania G, Della
`Mea V & Beltrami CA 1996 Large core biopsy for diagnostic
`and prognostic evaluation of invasive breast carcinomas.
`European Journal of Cancer 32A 1693-1700.
`Early Breast Cancer Trialists’ Collaborative Group 1998a
`Tamoxifen for early breast cancer: an overview of the
`randomised trials. Early Breast Cancer Trialists’ Collaborative
`Group. Lancet 351 1451-1467.
`Early Breast Cancer Trialists’ Collaborative Group 1998b
`Polychemotherapy for early breast cancer: an overview of the
`randomised trials. Early Breast Cancer Trialists’ Collaborative
`Group. Lancet 352 930-942.
`Fisher B, Bryant J, Wolmark N, Mamounas E, Brown A, Fisher
`ER, Wickerham DL, Begovic M, DeCillis A, Robidoux A,
`Margolese RG, Cruz AB Jr, Hoehn JL, Lees AW, Dimitrov
`NV & Bear HD 1998 Effect of preoperative chemotherapy on
`the outcome of women with operable breast cancer. Journal of
`Clinical Oncology 16 2672-2685.
`Frassoldati A, Adami F, Banzi C, Criscuolo M, Piccinini L &
`Silingardi V 1997 Changes of biological features in breast
`cancer cells determined by primary chemotherapy. Breast
`Cancer Research and Treatment 44 185-192.
`Harvey JM, Clark GM, Osborne CK & Allred DC 1999 Estrogen
`receptor status by immunohistochemistry is superior to the
`ligand—binding assay for predicting response to adjuvant
`endocrine therapy in breast cancer. Journal of Clinical Oncology
`17 1474-1481.
`Hawkins RA, Tesdale AL, Anderson ED, Levack PA,