Arimidex) on Intratumoral Estrogen Levels and Proliferation Markers in Patients with Locally Advanced Breast Cancer1

1230 Vol. 7, 1230–1236, May 2001 Clinical Cancer Research

Influence of Neoadjuvant Anastrozole (Arimidex) on Intratumoral Estrogen Levels and Proliferation Markers in Patients with Locally Advanced Breast Cancer1

Ju¨rgen Geisler, Simone Detre, expression of the proliferation markers Ki67 and pS2 in all Hildegunn Berntsen, Lars Ottestad, of the patients, with a trend for a more profound suppres- Bernt Lindtjørn, Mitch Dowsett, and sion in those achieving an objective response. The mean 2 percentage of apoptotic cells was found to be decreased in Per Eystein Lønning responders and increased in nonresponders after 15 weeks Departments of Oncology [J. G., H. B., P. E. L.], and Surgery [B. L.], of anastrozole therapy. Our results reveal anastrozole to Haukeland University Hospital, N-5021 Bergen, Norway; Academic cause a significant suppression of tissue estrogen levels and Department of Biochemistry, Royal Marsden Hospital, London, SW3 6JJ, United Kingdom [S. D., M. D.]; and Department of Oncology, to influence the biology of primary estrogen receptor-posi- The Norwegian Radiumhospital, N-0310 Oslo, Norway [L. O.] tive breast cancers in postmenopausal women.

INTRODUCTION ABSTRACT Aromatase inhibitors are successfully used for the therapy Anastrozole (Arimidex) is a novel, selective, and potent of postmenopausal breast cancer (1, 2). Although novel, third- aromatase inhibitor used for the treatment of postmeno- generation drugs like letrozole, anastrozole, and exemestane pausal breast cancer. The drug has been shown to inhibit in have been shown to cause profound suppression of plasma vivo aromatization by 96–97% and to suppress plasma es- estrogen levels and to inhibit total body aromatization by 96 to trogen levels by 84–94%. However, the effects of anastrozole Ͼ99% (3–5), these changes may not necessarily reflect alter- on intratumoral estrogen levels have not been studied. Here ations of tumor estrogen levels. Breast tumor tissue contains we report the effects of neoadjuvant treatment with anas- sulfatase as well as aromatase and dehydrogenase (6, 7), allow- trozole on intratumoral levels of estrone (E ), estradiol (E ), 3 1 2 ing synthesis of E from androgens as well as from E and E S. and estrone sulfate (E S), measured by a highly sensitive 2 1 1 1 Whereas the aromatase in tumor tissue is the same enzyme as RIA following a multistep purification procedure involving detected in other tissues, its expression may be enhanced by high-pressure liquid chromatography. Tumor tissue was ob- growth factors and interleukins locally synthesized (8). In ad- tained prior to treatment and after 15 weeks on therapy with dition, recent evidence has suggested active uptake of E from anastrozole (1 mg once daily) from 12 postmenopausal 2 the circulation in tumor cells (9). women with locally advanced breast cancer (T –T and/or 3 4 We have recently developed a highly sensitive and specific N ). Pretreatment tissue levels of E ,E , and E S were 217.9 2 2 1 1 HPLC-RIA method for the simultaneous measurement of E , (69.8–679.9), 173.6 (83.9–358.9), and 80.7 (31.4–207.3) 2 E , and E S levels in breast cancer tissue (10). The aim of this fmol/g tissue (geometric mean values with 95% confidence 1 1 study was to determine the influence of anastrozole treatment on interval, respectively). Treatment with anastrozole sup- intratumoral estrogen levels in postmenopausal women with pressed tissue E ,E , and E S levels by 89.0% (73.2–95.5%), 2 1 1 locally advanced breast cancer who had not been exposed to 83.4% (63.2–92.5%), and 72.9% (47.3–86.1%), respectively, previous anticancer therapy, by determining estrogen levels in compared with baseline levels, with no significant difference the same tumors prior to and after 15 weeks of neoadjuvant between responders and nonresponders. Plasma levels of E , 2 anastrozole therapy. To evaluate the biological effects of estro- E , and E S were suppressed by 86.1, 83.9, and 94.2%, 1 1 gen suppression, we measured changes in the proliferation respectively. Anastrozole caused a decrease in the immuno- marker Ki67 and the percentage of apoptotic cells as well as hormone and growth factor receptors (ER, PgR, c-erbB-2, EGF-R) and the estrogen-dependent protein pS2.

Received 9/27/00; revised 1/3/01; accepted 2/15/01. PATIENTS AND METHODS The costs of publication of this article were defrayed in part by the Patients. Postmenopausal women with locally advanced payment of page charges. This article must therefore be hereby marked (T ,T, and/or N ) noninflammatory breast cancer, with or advertisement in accordance with 18 U.S.C. Section 1734 solely to 3 4 2 indicate this fact. without limited distant metastasis, were eligible. Fourteen pa- 1 Supported by grants from the Norwegian Cancer Society, the Cancer tients fulfilled the inclusion criteria and started treatment ac- Research Campaign (Untied Kingdom), and Zeneca Pharmaceuticals (AstraZeneca). Parts of this study were presented at the American Society of Clinical Oncology meeting in Atlanta 1999 and the Notting- ham International Breast Cancer Conference 1999. Arimidex is a trade- 3 mark, the property of Zeneca Pharmaceuticals, a part of AstraZeneca. The abbreviations used are: E2, estradiol; E1, estrone; E1S, estrone 2 To whom requests for reprints should be addressed, at Department of sulfate; HPLC, high-performance liquid chromatography; CI, confi- Oncology, Haukeland University Hospital, N-5021 Bergen, Norway. dence interval; ER, estrogen receptor; PgR, progesterone receptor; Phone: 47-55-97-2010; Fax: 47-55-97-2046; E-mail: [email protected]. EGF-R, epidermal growth factor receptor; IHC, immunohistochemistry.

cording to protocol. However, one patient refused surgery after to RIA analysis. E1 and E1S were converted into E2, and all having a partial response and another patient died after only 2 three estrogen fractions were finally measured by the same weeks on treatment for reasons not related to her breast cancer highly sensitive and specific RIA using estradiol-6-carboxy- (cardiopulmonary insufficiency). The remaining 12 patients methyloxime-[2-125I]-iodohistamine as a ligand. Final estrogen (median age, 66.5 years; range 55–80) completed the study values were corrected for the amount of tissue used in each (Table 1). Postmenopausal status was defined as amenorrhea for individual sample (wet weight) as well as for the recovery of the Ն1 year duration with luteinizing hormone/follicle-stimulating amount of [3H]-hormone added as internal standard. The detec- hormone levels in the postmenopausal range. All of the tumors tion limits for tissue levels of E2,E1, and E1S were 4.3, 19.8, were ER positive (Ն10 fmol/mg tissue or Ն10% of the tumor and 11.9 fmol/g tissue, respectively (10). cells staining positive for ER by IHC) as determined prior to Collection of Blood Samples and Measurement of Plasma inclusion. To assess the influence of therapy on this parameter, Estrogen Levels. Blood samples for hormone measurements samples collected before and during treatment were redeter- were taken into heparinized vials (2 vials containing 10 ml each) mined with use of the same immunoassay as described in the immediately before commencing therapy with anastrozole and method section. Exclusion criteria included hormone replace- after 15 weeks on treatment. Each sample was obtained after ment therapy, systemic treatment with glucocorticoids, or an overnight fast. Plasma was separated by centrifugation and treatment with other drugs, like phenytoin, carbamazepine, or stored at Ϫ20°C until processing. E and E were determined rifampicin, that are known to influence estrogen metabolism 2 1 by RIAs as reported elsewhere (14, 15). Plasma levels of E1S (11, 12), within 3 months prior to protocol inclusion. One were determined by a novel, highly sensitive assay involving patient (no. 8) received treatment with glucocorticoid inha- purification and derivatization into E and by RIA analysis lations because of an obstructive lung disease. The patients 2 using estradiol-6-carboxymethyloxime-[2-125I]-iodohistamine as were treated at the Departments of Oncology, Haukeland tracer ligand (16). The sensitivity limits for plasma levels of E ,E , University Hospital, Bergen, and at The Norwegian Radium- 2 1 and E S were 2.1, 6.3, and 2.7 pmol/L, respectively. hospital, Oslo, Norway. 1 Tissue Biomarkers. All of the biomarkers were analyzed Treatment. Each patient received anastrozole (Arimi- by previously published methodology on histological sections. ER dex) as an oral dose of 1 mg once daily. In general, treatment expression was demonstrated with the DAKO 1D5 mouse mono- was given for 15 weeks, followed by surgery (mastectomy) and clonal (17) and PgR with the Novocastra antibody NCL-PgR clone radiation. Each breast tumor (and, if present, palpable axillary nodes), was measured prior to therapy and after 4, 8, 12, and 15 1A6 (18). Measurement of Ki67 was performed with the MIB1 weeks on anastrozole therapy. Tumor size was calculated as the mouse monoclonal antibody and of apoptosis by the TUNEL product of the largest diameter and its perpendicular. Clinical method (terminal deoxynucleotidyl transferase-mediated dUTP- response was classified according to the Union International biotin nick end labeling; Ref. 19). To ensure acceptable precision, Contre Cancer criteria (13), with the exception that tumors quantitation of apoptosis and Ki67 involved the counting of 3000 reduced by Ն25% but less than 50% in size were classified as and 1000 cells, respectively. For all other analytes, 10 high- “minimal change.” All of the patients who experienced a partial powered fields (chosen randomly) were assessed. Notably, the response or a minimal change were defined as responders, cellularity of the tissue samples obtained after treatment with anas- whereas all of the patients with stable disease or a progressive trozole did not vary substantially from pretreatment ones. c-erbB2 disease were defined as nonresponders. Patients who experi- and EGF-R were assessed with the ICR12 and Biogenex MU207 enced an objective response to treatment with anastrozole con- antibodies, respectively (20). pS2 antibody was a gift from tinued with anastrozole as adjuvant therapy for up to 5 years. Prof. Pierre Chambon (Institut de Genetique et de Biologie Molec- The protocol was approved by the local ethical committee, and ulaire et Cellulaire, CNRS/INSE Louis Pasteur, College de France, every patient gave her written informed consent. Strasbourg, France). Staining was scored as described previously Tissue Collection. Breast tumor tissue was collected except for ER and PgR, which were reported as percentage of prior to treatment (open biopsy), after 2–3 weeks on treatment positively stained cells. (true-cut biopsy), and during final surgery (mastectomy). All of Statistics. Plasma and tissue estrogen levels were described the tissue samples were stripped of adhering fat and were by their geometric means with 95% CI limits. Whenever estrogen divided into several pieces of about 100 mg each and one single levels (either in plasma or tissue samples) below the detection piece of about 500 mg. Tissue samples to be used for estrogen limits were found, the value of the respective detection limit was measurements were immediately snap-frozen and stored in liq- ascribed to the sample for statistical analysis. Before-treatment and uid nitrogen until analysis. For IHC, tissue samples were fixed on-treatment values were compared using the Wilcoxon matched- in formalin and embedded in paraffin. pair signed rank test. The mean value of percentage suppression Measurement of Intratumoral Estrogen Levels. Tis- from baseline for a parameter was calculated as 100 Ϫ x, where x sue estrogen concentrations were measured using a novel highly is the geometric mean value of the individual parameters in the sensitive RIA method subsequent to a multiple-step purification on-treatment situation expressed as percentage of pretreatment val- process involving HPLC as described elsewhere (10). Briefly, ues (4). All of the quantitative results generated by IHC (percentage tissue homogenates were incubated with [3H]-labeled estrogens of ER and PgR, Ki67, pS2 and percentage of apoptotic cells) are

(E1,E2,E1S) as recovery controls and crude fractions were given as their arithmetric mean levels. EGF-R and c-erbB-2 status separated by ether extraction. The E1S fraction was hydrolyzed were described as either positive or negative. The Friedman test with sulfatase followed by elution on a Sephadex column. was used to test for differences between three groups (see Table 2 HPLC was used to purify the individual estrogen fractions prior for details). The comparison of changes in proliferation markers

and receptor levels between responders and nonresponders was performed using the Mann-Whitney U test.

RESULTS

The mean tissue concentrations of E2,E1, and E1Sat baseline were 217.9 fmol/g (95% CI, 69.8–679.9 fmol/g), 173.6 fmol/g (95% CI, 83.9–358.9 fmol/g), and 80.7 fmol/g (95% CI, 31.4–207.3 fmol/g), respectively. There was a nonsignificant trend for a positive correlation (r-0.40) between the percentage

of ER positive cells and E2 levels in the tumor tissues. Treatment with anastrozole suppressed tissue concentra-

tions of E2,E1, and E1S (Fig. 1) to 23.9 fmol/g (95% CI, 12.2–47.0 fmol/g), 28.8 fmol/g (95% CI, 19.7–42.1 fmol/g), and 21.9 fmol/g (95% CI, 12.6–37.9 fmol/g). The percentage

suppression of E2,E1, and E1S were 89.0% (95% CI, 73.2– 95.5%), 83.4% (95% CI, 63.2–92.5%), and 72.9% (95% CI, 47.3–86.1%), respectively (Fig. 2). It should be noted that one,

eight, and seven patients, respectively, had tissue levels of E2,

E1, and E1S below the detection limit during treatment. There was no significant correlation between the tissue concentrations of any estrogen fraction before or after therapy with anastrozole.

Pretreatment plasma levels of E2,E1, and E1S were 18.4 pmol/liter (95% CI, 13.2–25.6 pmol/liter), 70.6 pmol/liter (95% CI, 55.3–90.1 pmol/liter), and 578.8 pmol/liter (95% CI, 328.7– 1020.5 pmol/liter), respectively. After 15 weeks on anastrozole

therapy the plasma levels of E2,E1, and E1S decreased to 2.6 pmol/liter (95% CI, 2.1–3.1 pmol/liter), 11.4 pmol/liter (95% CI, 9.5–13.7 pmol/liter), and 33.3 pmol/liter (95% CI, 16.6– 67.0 pmol/liter), which corresponded to a suppression of 86.1% (95% CI, 81.6–89.5%), 83.9% (95% CI, 79.0–87.6%), and 94.2% (95% CI, 90.7–96.5%), respectively. Seven patients had

plasma levels of E2, and one patient had plasma levels of E1S below the detection limit while on treatment with anastrozole. The patient (no. 8) receiving glucocorticoids by inhalation had

low pretreatment plasma estrogen levels with values of E2,E1,

and E1S of 8.4, 32.2,, and 330.9 pmol/liter, respectively. Patient demographics, clinical responses, and individual expression of growth factor receptors are given in Table 1. Whereas true-cut biopsies were obtained from all of the patients after 2–3 weeks on treatment, IHC was evaluable for seven to eight patients only, because amounts of tissue in some biopsies were too small. However, open biopsies were available for all patients before treatment and surgical specimens (obtained di- rectly from the removed breast) after 15 weeks (median) on treatment with anastrozole. There was a trend for ER levels to be higher in responders than in nonresponders (Tables 1 and 2). Whereas expression of ER did not change during therapy, PgR levels fell significantly. The suppression was similar in responders and nonresponders. A similar pattern of change was seen for pS2, except that in this case, the fall was already significant after 2 weeks on therapy. For the proliferation marker Ki67, there was a significant decrease at 2–3 weeks and an additional decrease to 37% of baseline values at 3 months. This was quantitatively greater for responders but statistically significant for both responders and Fig. 1 A–C, influence of treatment with neoadjuvant anastrozole on nonresponders. Overall, there was no significant change in the intratissue estrogen (fmol/g tissue) levels (F, responders; E, nonre- mean percentage of apoptotic cells. However, it decreased in sponders).

29–35), only a few studies have determined the effect of aromatase inhibitor therapy on intratumoral estrogen levels. Reed et al. (36) investigated the influence of the second-generation steroidal aromatase inhibitor 4-hydroxyandrostenedione on tu-

mor E1 levels in four patients and reported a reduction by 62%. de Jong and colleagues measured intratumoral E1 and E2 levels in five patients on treatment with vorozole, an aromatase inhibitor belonging to the triazole class (37). Because of the lack of pretreatment levels, they compared their results with historical

data for untreated patients, suggesting a suppression of tissue E1 and E2 levels by 64% and 80%, respectively. Although Miller et Fig. 2 Percentage of pretreatment levels of E ,E, and E S in plasma 1 2 1 al. (38) determined tumor E1 and E2 levels in patients before and (P) and tissue (T) samples during treatment with anastrozole. during treatment with letrozole, the exact percentage of suppression could not be determined because of the concomitant infu- sion of tracer steroids used to determine estrogen production rates. ϭ responders and increased in nonresponders (P 0.045, when The present study revealed profound suppression of tu- comparing the two groups). moral E1 and E2 levels to an extent similar to that of plasma Two nonresponding tumors showed positive staining for estrogen suppression. The exact proportional decreases were not c-erbB-2; one of these tumors, in addition, expressed EGF-R. It possible to calculate because a number of patients had levels was notable that the two patients expressing c-erbB-2 also had suppressed below the detection limit of even these highly sen- the lowest ER levels and the lowest tissue E2 levels of all of the sitive assays. The finding of a better suppression of tissue E1 and participants in this trial. No change in the expression of either of E2 levels compared with tissue E1S levels may also be partly these receptors was observed during therapy with anastrozole. In explained by three of the patients having tissue E1S levels below addition, no significant correlation was found between staining the detection limit at baseline. In addition, 7 of the 12 patients for c-erbB-2, ER levels, estrogen tissue or plasma levels and had on-treatment tissue E1S levels below the detection limit; clinical response in the whole study group. thus, an underestimation of the E1S suppression is likely in the descriptive statistics. It is notable that patients with high intra-

DISCUSSION tumoral E1 and E2 levels experienced, in general, a better Although several studies have determined plasma estrogen estrogen suppression compared with those with lower levels at levels and total body aromatization in patients treated with novel baseline. aromatase inhibitors (3–5), there is limited information about In marked contrast to a previous study (22), which reported alterations in tumor estrogen concentrations during therapy. It is tissue E1S levels to be 10-fold higher than plasma levels, we ϳ well documented that breast cancer E2 concentrations are 10- to found tissue levels of E1S to be only 20% of plasma E1S 20-fold higher and E1 concentrations 2- to 10-fold higher than levels. E1S is the most abundant circulating estrogen in post- the corresponding plasma levels in postmenopausal women (21– menopausal women and conversion of E1StoE1 and E2 has 24), a finding confirmed in this study. Whether this may be been suggested to be a major pathway of intratumoral estrogen attributable to local synthesis (6) or active estrogen uptake from synthesis (6, 39). Because of its hydrophilic nature, E1Sis the circulation (9) is unknown. Studies published by Miller (25) unlikely to pass the cell membrane. However, after hydrolysis of and by James et al. (26), evaluating the origin of intratumoral E1S by the ubiquitous sulfatase, E1 may pass through cell estrogens, reported a substantial interindividual variation, which membranes and be converted to either E2 or E1S in the cells. suggested that the bulk of intratumoral estrogens occur from Thus, whereas our data are consistent with a lack of uptake of local synthesis in some tumors but are derived from the circu- E1S per se and refute major conversion of E1 to E1S in tumors, lation in others. Although the aromatase enzyme is the same in our findings do not refute the possibility that circulating E1S both malignant and normal cells, the local activity of the enzyme may contribute to intratumoral estrogens after deconjugation may be influenced by growth factors and cytokines (27, 28). prior to uptake. Thus, direct measurements are necessary to evaluate the influ- We found no correlation between the suppression of any of ence of different treatment modalities on intratumoral estrogens. the tissue estrogen fractions and the clinical response to anas- A key problem associated with intratumoral estrogen meas- trozole treatment. Thus, our data do not suggest that resistance urements during treatment with aromatase inhibitors is the re- to neoadjuvant treatment with anastrozole can be explained by quirement of a detection method with sensitivity limits suitable an incomplete suppression of tissue estrogens; intrinsic resist- for estrogen measurements in the low range. We have recently ance of the tumor to estrogen deprivation therapy is a more developed a novel, highly sensitive, and specific HPLC-RIA for probable explanation. the simultaneous measurement of E2,E1, and E1S in breast Several studies have investigated biomarkers that are po- tissue biopsies (10). The sensitivity limits of this method allow tentially predictive for response to neoadjuvant endocrine treat- the measurement of estrogen levels in the range expected during ment in addition to pretreatment ER and PgR levels. The ma- treatment of postmenopausal women with highly potent aro- jority of these studies have involved tamoxifen therapy. The matase inhibitors. Although several investigators have measured numbers of patients in the present study were insufficient to intratumoral estrogen levels in untreated patients (7, 21–24, provide data of substantial statistical strength, but these findings

Table 1 Patients’ characteristics Patient Age BMIa TNM % ERb % PgRb Met. Resp. c-erbB-2c EGF-Rc d 1 56 28.0 T3N0M0 2.0 0.0 PD pos./pos. pos./pos. 2 73 22.4 T4N2M1 98.0 0.0 bone MC neg./neg. neg./neg. 3 67 30.8 T4N2M1 86.0 53.0 bone PR neg./neg. neg./neg. 4 67 31.9 T3N1M0 92.0 86.0 MC neg./neg. neg./neg. 5 55 30.5 T3N2M0 87.0 70.0 MC neg./neg. neg./neg. 6 70 29.9 T2N2M0 82.0 7.5 PD neg./neg. neg./neg. d 7 58 26.4 T3N0M0 0.0 4.3 StD pos./pos. neg./neg. 8 70 29.2 T4N0M1 93.0 0.0 skin PR neg./neg. neg./neg. 9 66 25.8 T3N1M0 92.0 79.0 StD neg./neg. neg./neg. 10 80 26.9 T3N1M0 91.0 86.0 StD neg./neg. neg./neg. d 11 60 25.4 T4N1M0 n.a. n.a. StD n.a./neg. n.a./neg. 12 65 22.0 T4N0M0 83.0 84.0 PR neg./neg. neg./neg. a BMI, body mass index; TNM, tumor-node-metastasis; Met., metastasis; Resp., response; PD, progressive disease; pos., positive; MC, minimal change; neg., negative; PR, partial response; StD, stable disease; n.a., not available. b Expressed as percentage of cells staining positively (IHC). c Staining for c-erbB-2 and EGF-R is given prior to treatment with anastrozole (x/) and after 15 weeks (median) on treatment (/x). d ER levels Ն 10 fmol/mg were confirmed by the charcoal method prior to treatment with anastrozole.

Table 2 Influences of neoadjuvant Arimidex on ER, PgR, Ki67, pS2, and percentage of apoptotic cells (IHC) All of the values are given as arithmetric means. Pretreatment levels were compared to on-treatment levels after 2 weeks (0/2 wk) and 15 weeks (0/15 wk) using the Wilcoxon matched-pair signed rank test, and the Friedman test was used to test for differences among all three situations. Comparisons between subgroups of responders and nonresponders were performed using the Mann-Whitney U test. IHC data are not given for subgroups after 2 weeks on treatment with anastrozole because of the small number of evaluable true-cut biopsies (n ϭ 7–8). Arimidex Arimidex % change P for change P for change P between Pretreatment (2 wk) (15 wk) (0/15 wk) Friedman test (0/2 wk) (0/15 wk) subgroups ER (%)a All patients 73.3 65.9 71.6 Ϫ3.4% 0.1210 0.0172 n.s.b n.s. Responders 89.8 86.2 Ϫ4.1% n.s. Nonresponders 53.4 57.0 ϩ2.0% n.s. PgR (%) All patients 42.7 27.3 5.3 Ϫ63.9% 0.0119 0.0625 0.0117 n.s. Responders 48.8 0.0 Ϫ66.7% 0.0679 Nonresponders 35.4 10.6 Ϫ60.7% 0.0679 Ki67 All patients 17.3 11.3 7.7 Ϫ63.1% 0.0007 0.0322 0.0033 n.s. (0.0679) Responders 16.9 3.3 Ϫ82.8% 0.0277 Nonresponders 17.8 12.2 Ϫ39.5% 0.0431 pS2 All patients 43.6 3.2 16.1 Ϫ71.7% 0.0002 0.0051 0.0033 n.s. Responders 50.2 12.5 Ϫ77.6% 0.0277 Nonresponders 35.8 19.6 Ϫ64.7% 0.0431 Apoptolic cells All patients 1.3 1.2 1.5 Ϫ23.2% 0.4966 0.3863 n.s. 0.0446 Responders 1.1 0.7 Ϫ42.9% n.s. (0.0747) Nonresponders 1.6 2.3 ϩ0.5% n.s. a ER (%), percentage of tumor cells staining positive. b n.s., not significant.

remain of interest because they are some of the first to be that estrogen deprivation in the MCF7 human xenograft model described in terms of the effects of an aromatase inhibitor. leads to increased ER expression. The reason for this discrep- As expected, the two patients (nos. 1 and 7) who were ancy in the human breast cancer data is unclear. found to have low levels of ER determined biochemically (10.4 We found a marked decrease of PgR and pS2 expression and 11.1 fmol/mg, respectively), but who were later found to be during treatment. This is consistent with the known estrogen- negative by immunostaining, did not respond to therapy. These induced synthesis of these proteins (40, 41). Although our two patients showed expression of the type I growth factor finding that PgR and pS2 were suppressed to the same extent in receptors, EGF-R and/or c-erbB2, which is also consistent with responders and nonresponders contrasts with observations dur- the known inverse correlation between these receptors and ER. ing treatment with tamoxifen (42), the suppression of Ki67 The lack of effect of anastrozole treatment on ER levels is seems to be more profound in responders compared with non- somewhat surprising, because it is known that estrogen down- responders on both treatments. regulates its own receptor, and we have previously shown (18) The finding of a decrease in the number of apoptotic cells

among responders but an increase in nonresponders to treatment trations in the absence of ovarian function: role of very high affinity during anastrozole treatment may seem contradictory. However, tissue uptake. Breast Cancer Res. Treat., 42: 215–226, 1997. a high mitotic activity has been associated with a high apoptotic 10. Geisler, J., Berntsen, H., and Lønning, P. E. A novel HPLC-RIA activity in invasive breast cancer (43). Thus, our finding may method for the simultaneous detection of estrone, estradiol, and estrone sulphate levels in breast cancer tissue. J. Steroid Biochem. Mol. Biol., indicate that anastrozole decreases the cell turnover in tumors 72: 259–264, 2000. responding to therapy. Our observation is in contrast to the 11. Geisler, J., Engelsen, B., Berntsen, H., Geisler, S., and Lønning, findings made with neoadjuvant chemotherapy in which, re- P. E. Differential influence of carbamazepine and valproate mono- sponses are in general associated with an induction of apoptosis therapy on plasma levels of oestrone sulphate and dehydroepiandros- (19) and with the minor increase in apoptosis seen with anties- terone sulphate in male epileptic patients. J. Endocrinol., 153: 307–312, trogens after 1–2 weeks treatment (44). Recent data from the use 1997. of vorozole, another aromatase inhibitor, also showed a de- 12. Lønning, P. E., Bakke, P., Thorsen, T., Olsen, B., and Gulsvik, A. creased apoptotic index after 2 weeks treatment (45). Thus, Plasma levels of estradiol, estrone, estrone sulfate, and sex hormone binding globulin in patients receiving rifampicin. J. Steroid Biochem., effects on apoptosis may be differential between antiestrogens 33: 631–635, 1989. and aromatase inhibitors. 13. Hayward, J. L., Rubens, R. D., Carbone, P. P., Heuson, J-C., In conclusion, we found that anastrozole given as neoad- Kumaoka, S., and Segaloff, A. Assessment of response to therapy in juvant therapy causes a profound suppression of tumor levels of advanced breast cancer. Br. J. Cancer, 35: 292–298, 1977.

E2,E1, and E1S, and that this is associated with a reduction of 14. Dowsett, M., Goss, P. E., Powles, T. J., Hutchinson, G., Brodie, tumor cell proliferation and of the estrogen-regulated proteins A. M. H., Jeffcoate, S. L., and Coombes, R. C. Use of the aromatase PgR and pS2. inhibitor 4-hydroxyandrostenedione in postmenopausal breast cancer: optimization of therapeutic dose and route. Cancer Res., 47: 1957–1961, 1987. ACKNOWLEDGMENTS 15. Lønning, P. E., Helle, S. I., Johannessen, D. C., Adlercreutz, H., We thank the skillful technical assistance of Dagfinn Ekse in Lien, E. A., Tally, M., Ekse, D., Fotsis, T., Anker, G. B., and Hall, K. determining plasma hormone levels. Relations between sex hormones, sex hormone binding globulin, insulin-like growth factor-I, and insulin-like growth factor binding protein-1 in post-menopausal breast cancer patients. Clin. Endocrinol., 42: 23–30, REFERENCES 1995. 1. Buzdar, A., Jonat, W., Howell, A., Jones, S. E., Blomquist, C., 16. Lønning, P. E., and Ekse, D. A sensitive assay for measurement of Vogel, C. L., Eiermann, W., Wolter, J. M., Azab, M., Webster, A., and plasma estrone sulphate in patients on treatment with aromatase inhib- Plourde, P. V. Anastrozole, a potent and selective aromatase inhibitor, itors J. Steroid Biochem. Mol. Biol., 55: 409–412, 1995. versus megestrol acetate in postmenopausal women with advanced 17. Saccani-Jotti, G., Johnston, S. R., Salter, J., Detre, S., and Dowsett, breast cancer: results of overview analysis of two Phase III trials. J. Clin. Oncol., 14: 2000–2011, 1996. M. Comparison of new immunohistochemical assay for oestrogen receptor in paraffin wax embedded breast carcinoma tissue with quanti- 2. Dombernowsky, P., Smith, I., Falkson, G., Leonhard, R., Panasci, L., tative enzyme immunoassay. J. Clin. Pathol., 47: 900–905, 1994. Bellmunt, J., Bezwoda, W., Gardin, G., Gudgeon, A., Morgan, M., Fornasiero, A., Hoffmann, W., Michel, J., Hatschek, T., Tjabbes, T., 18. Detre, S., Salter, J., Barnes, D. M., Riddler, S., Hills, M., Johnston, Chaudri, H. A., Hornberger, U., and Trunet, P. F. Letrozole, a new oral S. R. D., Gillett, C., A’Hern, R., and Dowsett, M. Time-related effects aromatase inhibitor for advanced breast cancer: double-blind random- of estrogen withdrawal on proliferation- and cell death-related events in ized trial showing a dose effect and improved efficacy and tolerability MCF-7 xenografts. Int. J. Cancer, 81: 309–313, 1999. compared to megestrole acetate. J. Clin. Oncol., 16: 453–461, 1998. 19. Ellis, P. A., Smith, I. E., Detre, S., Burton, S. A., Salter, J., A’Hern, 3. Dowsett, M., Jones, A., Johnston, S. R. D., Jacobs, S., Trunet, P., and R., Walsh, G., Johnston, S. R. D., and Dowsett, M. Reduced apoptosis Smith, I. E. In vivo measurement of aromatase inhibition by letrozole and proliferation and increased bcl-2 in residual breast cancer following (CGS 20267) in postmenopausal patients with breast cancer. Clin. preoperative chemotherapy. Breast Cancer Res. Treat., 48: 107–116, Cancer Res., 1: 1511–1515, 1995. 1998. 4. Geisler, J., King, N., Dowsett, M., Ottestad, L., Lundgren, S., Wal- 20. Newby, J. C., Johnston, S. R. D., Smith, I. E., and Dowsett, M. ton, P., Kormeset, P. O., and Lønning, P. E. Influence of anastrozole Expression of epidermal growth factor receptor and c-erbB-2 during the (Arimidex), a selective, nonsteroidal aromatase inhibitor, on in vivo development of tamoxifen resistance in human breast cancer. Clin. aromatisation and plasma oestrogen levels in postmenopausal women Cancer Res., 3: 1643–1651, 1997. with breast cancer. Br. J. Cancer, 74: 1286–1291, 1996. 21. Fishman, J., Nisselbaum, J. S., Menendez-Botet, C. J., and 5. Geisler, J., King, N., Anker, G., Ornati, G., Salle, E. D., Lønning, Schwartz, M. K. Estrone and estradiol content in human breast tumors: P. E., and Dowsett, M. In vivo inhibition of aromatization by exemes- relationship to estradiol receptors. J. Steroid Biochem., 8: 893–896, tane, a novel irreversible aromatase inhibitor, in postmenopausal breast 1977. cancer patients. Clin. Cancer Res., 4: 2089–2093, 1998. 22. Pasqualini, J. R., Chetrite, G., Blacker, C., Feinstein, M-C., Dela- 6. Santner, S. J., Feil, P. D., and Santen, R. J. In situ estrogen produc- londe, L., Talbi, M., and Maloche, C. Concentrations of estrone, estra- tion via the estrone sulfatase pathway in breast tumors: relative impor- diol, and estrone sulfate and evaluation of sulfatase and aromatase tance versus the aromatase pathway. J. Clin. Endocrinol. Metab., 59: activities in pre- and postmenopausal breast cancer patients. J. Clin. 29–33, 1984. Endocrinol. Metab., 81: 1460–1464, 1996. 7. Bonney, R. C., Reed, M. J., Davidson, K., Beranek, P. A., and James, 23. Vermeulen, A., Deslypere, J. P., Paridaens, R., Leclercq, G., Roy, V. H. T. The relationship between 17-␤-hydroxysteroid dehydrogenase F., and Heuson, J. C. Aromatase, 17-␤-hydroxysteroid dehydrogenase activity and oestrogen concentrations in human breast tumours and in and intratissue sex hormone concentrations in cancerous and normal normal breast tissue. Clin. Endocrinol., 19: 727–739, 1983. glandular breast tissue in postmenopausal women. Eur. J. Cancer Clin. 8. Reed, M. J., and Purohit, A. Breast cancer and the role of cytokines Oncol., 22: 515–525, 1986. in regulating estrogen synthesis: an emerging hypothesis. Endocr. Rev., 24. Landeghem, A. A. J. v., Poortman, J., Nabuurs, M., and Thijssen, 18: 701–715, 1997. J. H. H. Endogenous concentration and subcellular distribution of es- 9. Masamura, S., Santner, S. J., Gimotty, P., George, J., and Santen, trogens in normal and malignant breast tissue Cancer Res., 45: 2900– R. J. Mechanism of maintenance of high breast tumor estradiol concen- 2906, 1985.

25. Miller, W. R. Importance of intratumour aromatase and its suscep- one in malignant and normal breast tissue. Int. J. Cancer, 49: 562–565, tibility to inhibitors. In: M. Dowsett (ed.), Aromatase Inhibition—Then, 1991. Now, and Tomorrow. London: Parthenon Publishing Group, 1994. 37. de Jong, P. C., van de Ven, J., Nortier, H. W. R., Maitimu-Smeele, 26. James, V. H. T., Reed, M. J., Adams, E. F., Ghilchick, M., Lai, I., Donker, T. H., Thijssen, J. H. H., Slee, P. H. T. J., and Blankenstein, L. C., Coldham, N. G., Newton, C. J., Purohit, A., Owen, A. M., Singh, R. A. Inhibition of breast cancer tissue aromatase activity and estrogen A., and Islam, S. Oestrogen uptake and metabolism in vivo. In: J. S. concentrations by the third-generation aromatase inhibitor vorozole. Beck (ed.), Oestrogen and the Human Breast, Vol. 95B, pp. 185–193. Cancer Res., 57: 2109–2111, 1997. Edinburgh: The Royal Society of Edinburgh, 1989. 38. Miller, W. R., Telford, J., Love, C., Leonard, R. C. F., Hillier, S., 27. Reed, M. J., Topping, L., Coldham, N. G., Purohit, A., Ghilchik, Grundacker, H., Smith, H., and Dixon, J. M. Effects of letrozole as M. W., and James, V. H. T. Control of aromatase activity in breast primary medical therapy on in situ oestrogen synthesis and endogenous cancer cells: the role of cytokines and growth factors. J. Steroid Bio- oestrogen levels within the breast. Breast, 7: 273–276, 1998. chem. Mol. Biol., 44: 589–596, 1993. 39. Santner, S. J., Leszcynski, D., Wright, C., Manni, A., Feil, P. D., 28. Bulun, S. E., and Simpson, E. R. Regulation of aromatase expres- and Santen, R. J. Estrone sulfate: a potential source of estradiol in sion in human tissues. Breast Cancer Res. Treat., 30: 19–29, 1994. human breast cancer tissue. Breast Cancer Res. Treat., 7: 35–44, 1986. 29. Thorsen, T., Tangen, M., and Støa, K. F. Concentration of endog- 40. Horwitz, K. B., McGuire, W. L., Pearson, O. H., and Segaloff, A. enous oestradiol as related to oestradiol receptor sites in breast tumor Predicting response to endocrine therapy in human breast cancer: a hypothesis. Science (Wash. DC)., 189: 726–727, 1975. cytosol. Eur. J. Cancer Clin. Oncol., 18: 333–337, 1982. 41. Masiekowski, P., Breathnach, R., Bloch, J., Gannon, K., Krust, A., 30. Mistry, P., Griffith, K., and Maynard, P. W. Endogenous C19- and Chambon, P. Cloning of cDNA sequences of hormone-regulated steroids and oestradiol levels in human primary breast tumour tissues genes from MCF-7 human breast cancer cell line. Nucleic Acids Res., and their correlation with androgen and oestrogen receptors. J. Steroid 10: 7895–7903, 1982. Biochem., 24: 1117–1125, 1986. 42. Makris, A., Powles, T. J., Allred, D. C., Ashley, S., Ormerud, 31. Edery, M., Goussard, J., Dehennin, L., Scholler, R., Reiffsteck, J., ␤ M. G., Titley, J. C., and Dowsett, M. Changes in hormone receptors and and Drosdowsky, M. A. Endogenous oestradiol-17- concentration in proliferation markers in tamoxifen treated breast cancer patients and the breast tumours determined by mass fragmentography and by radioim- relationship with response. Breast Cancer Res. Treat., 48: 11–20, 1998. munoassay: relationship to receptor content. Eur. J. Cancer, 17: 115– 120, 1980. 43. v. Slooten, H-J., v. d. Vijver, M. J., Børresen, A-L., Eyfjo¨rd, J. E., Valgardsdottir, R., Scherneck, S., Nesland, J. M., Deville, P., Cornel- 32. Recchione, C., Venturelli, E., Manzari, A., Cavalleri, A., Martinetti, isse, C. J., and v. Dierendonk, J. H. Mutations in exons 5–8 of the p53 A., and Secreto, G. Testosterone, dihydrotestosterone, and oestradiol gene, independent of their type and localisation, are associated with levels in postmenopausal breast cancer tissues. J. Steroid Biochem. Mol. increased apoptosis and mitosis in invasive breast carcinoma. J. Pathol., Biol., 52: 541–546, 1995. 189: 504–513, 1999. 33. Millington, D. S. Determination of hormonal steroid concentrations 44. Ellis, P. A., Sacconi-Jotti, G., Clarke, R., Johnston, S. R. D., in biological extracts by high resolution mass fragmentography. J. Anderson, E., Howell, A., A’Hern, R., Salter, J., Detre, S., Nicholson, Steroid Biochem., 6: 239–245, 1975. R., Robertson, J., Smith, I. E., and Dowsett, M. Induction of apoptosis 34. Maynard, P. V., Brownsey, B. G., and Griffith, K. Oestradiol levels by tamoxifen and ICI 182780 in primary breast cancer. Int. J. Cancer, in fractions of human breast tumours. J. Endocrinol., 77: P62–P63, 72: 608–613, 1997. 1978. 45. Harper-Wynne, C., Shenton, K., Dowsett, M., MacNeil, F., Sauven, 35. Vallent, K., Feher, T., Bodrogi, L., and Ribai, Z. Steroid-hormon- P., Laidlaw, I., Rayter, Z., Miall, S., and Sacks, N. A randomised gehalt der brustsubstanz bei mammatumoren. Chirurg, 53. 34–36, 1982. multicentre study of vorozole compared to tamoxifen as primary therapy 36. Reed, M. J., Aherne, G. W., Ghilchik, M. W., Patel, S., and in post-menopausal breast cancer. Proc. Am. Soc. Clin. Oncol., abs. 272 Chakraborty, J. Concentrations of oestrone and 4-hydroxyandrostenedi- (pg. 72a), 1999.

Downloaded from clincancerres.aacrjournals.org on September 29, 2021. © 2001 American Association for Cancer Research. Influence of Neoadjuvant Anastrozole (Arimidex) on Intratumoral Estrogen Levels and Proliferation Markers in Patients with Locally Advanced Breast Cancer

Jürgen Geisler, Simone Detre, Hildegunn Berntsen, et al.

Clin Cancer Res 2001;7:1230-1236.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/5/1230

Cited articles This article cites 41 articles, 11 of which you can access for free at: http://clincancerres.aacrjournals.org/content/7/5/1230.full#ref-list-1

Citing articles This article has been cited by 18 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/7/5/1230.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/7/5/1230. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.