The Role of Tamoxifen and Aromatase Inhibitors/Inactivators in Postmenopausal Patients 1

4356s Vol. 7, 4356s-4359s, December 2001 (Suppl.) Clinical Cancer Research

Kathleen I. Pritchard in particular diethylstilbestrol, were used as first-line therapy in Toronto-Sunnybrook Regional Cancer Centre, University of Toronto, the treatment of postmenopausal women with metastatic breast Toronto, Ontario M4N 3M5, Canada cancer. This treatment was initially tried on all postmenopausal women, but with the elucidation of its mechanism of action by way of the ER (1), 3 physicians began to select patients for Abstract endocrine therapy on this basis. It has been shown repeatedly The traditional hormonal cascade of the 1970s and that women whose tumors overexpress ER and/or PgR are more 1980s used tamoxifen followed by megestrol acetate and likely to respond to endocrine therapies of any type. In partic- subsequently by aminoglutethimide. In the 1990s, however, ular, women whose tumors are positive for ER or PgR have three trials of third-generation aromatase inhibitors (AIs) response rates of --30%, while those whose tumors have high compared with megestrol acetate and two trials of third- levels of both receptors may have response rates as high as 60% generation AIs compared with aminoglutethimide showed or 70% to any hormonal approach. Higher levels of ER and PgR improved efficacy and decreased toxicity for the newer AIs. are closely correlated to response (2). Thus, the hormonal cascade changed in the late 1990s, to one In the 1970s and 80s the traditional hormonal cascade for in which tamoxifen, followed by a third-generation AI, fol- postmenopausat women involved tamoxifen followed by a pro- lowed by megestrol acetate, seemed more suitable. Now, gestational agent such as megestrol acetate (Megace) or however, several trials comparing anastrozole, letrozole, and medroxyprogesterone acetate (Provera), and subsequently ami- exemestane to tamoxifen as first-line hormonal agents for noglutethimide, the prototype aromatase inhibitor. More re- metastatic breast cancer have shown that these drugs are at cently, however, the results of a number of trials of third- least equivalent and perhaps superior to tamoxifen in that generation AIs compared with previous standard therapy in setting in terms of response rate and time to progression. second-line and now in first-line treatment of metastatic disease Results from 1021 patients randomized to receive anastro- have changed this approach dramatically. zole versus tamoxifen showed a slightly improved overall response rate (RR; 29% versus 26%), slightly improved clinical benefit (CB; 57% versus 52%), and a significantly The AIs and Their Historical Development improved time to progression (TTP; 8.5 months versus 7.0 Aminoglutethimide, the prototype AI, was originally de- months) in favor of anastrozole. In 907 women randomized veloped as an anticonvulsant. It was first identified as an inhib- to treatment with letrozole versus tamoxifen, significantly itor of adrenal steroidogenesis in 1967 and first used as a form improved RR (30% versus 20%), CB (49% versus 38%), and of medical adrenalectomy in 1970. Aminoglutethimide was TTP (9.4 months versus 6 months) have all been shown for introduced into clinical practice for advanced breast cancer in those treated with |etrozole. In addition, a randomized 1981. This drug, like the second- and third-generation AIs, acts Phase II trial of 121 patients has shown nonsignificant ben- by inhibiting the aromatase enzyme, the key enzyme mediating efits in favor of exemestane (RR 41% versus 14%; CB 56% the transformation of androstendione and testosterone to estrone versus 42 %; TTP not available). To date, none of these trials and estradiol. Since this aromatization process is the main has demonstrated any overall survival benefit. Additional source of estrogen in postmenopausal women, aromatase inhi- follow-up in regard to survival in the trial of tamoxifen bition dramatically reduces estrogen levels in such patients. versus letrozole and an expanded Phase III trial of tamoxifen Aminoglutethimide was in many ways a poor AI for clin- versus exemestane are ongoing. ical use in that it is not particularly specific. It inhibited not only aromatase but also the 17, 11, and 21 hydroxylation pathways of Introduction adrenal steroidogenesis, resulting in reduced levels of the glu- Hormonal therapy remains a mainstay in the treatment of cocorticoids and mineralocorticoids necessary for healthy life. women with metastatic breast cancer. However, the use of Thus, aminoglutethimide given in doses of more than 250-500 hormonal agents for metastatic breast cancer in postmenopausal mg daily required supplemental prednisone and/or fludrocorti- women deserves re-examination at this time. Initially, estrogens, sone in order to avoid adrenocortical insufficiency syndromes. The second-generation AIs, such as formestane, were the next important development. Formestane is a much more specific AI,

1 Presented at the First International Conference on Recent Advances and Future Directions in Endocrine Therapy for Breast Cancer, June 21-23, 2001, Cambridge, MA. 2 To whom requests for reprints should be addressed, at Toronto- 3 The abbreviations used are: ER, estrogen receptor; PgR, progesterone Sunnybrook Regional Cancer Centre, University of Toronto, 2075 Bay- receptor; OS, overall survival; TTF, time to treatment failure; AI, view Avenue, Toronto, Ontario M4N 3M5, Canada. Phone: (416)480- aromatase inhibitor; RR, response rate; CB, clinical benefit; TTP, time 4616; Fax: (416) 480-6002; E-mail: [email protected]. to progression.

Table 1 Metastatic breast cancer optimal endocrine therapy in postmenopausal women No. of trials showing superior outcome for new A.I.

3 trials of new A.I. O.S. TTF TTP RR challenging megace 2 3* 2* 1 7 N = 2250 (A/E) (L/E) (E) (L) Tamoxifen Failures 2 trials of new A.I. 1 2 1 challenging (L) (L/V) (L) aminoglutethimide N=919 A = Anastrozole L = Letrozole V = Vorozole E = Exemestane * improvement significant for letrozole .5 mg compared to megestrol acetate but not for letrozole 2.5 mg in reference 7.

but because it was available only in an injectable form, it was Trials of Third-Generation AIs in First-Line not the ideal AI for clinical use. Therapy for Metastatic Disease Subsequently, two groups of third-generation AIs were There are essentially three trials in which women with ER- developed. These include the nonsteroidal inhibitors anastrozole and/or PgR-positive or ER- and PgR-unknown tumors, disease- (Arimidex), letrozole (Femara), and vorozole (no longer in free interval > 1 year, and an adjuvant tamoxifen-free interval of commercial development), as well as the steroidal AI exemes- 6-12 months or greater, have been randomized after relapse to tane (Aromasin). The nonsteroidal inhibitors bind to the heme tamoxifen versus a third-generation AI (anastrozole, letrozole, part of the aromatase enzyme in a reversible fashion, while the or exemestane). steroidal inhibitors bind to the substrate binding site of the The trial of anastrozole versus tamoxifen actually consists enzyme in an irreversible fashion and are, therefore, often of two large, randomized trials of identical design, one carried referred to as aromatase inactivators (3). out in North America and one in Europe. These trials were designed to be analyzed in combination. In the two trials com- Trials of Third-Generation Als in Second-/Third- bined, 1021 patients were randomized to receive either anastro- Line Therapy for Metastatic Disease zole or tamoxifen (10, 11). The letrozole trial was run as one There are now six randomized trials of the third-generation large trial in which 907 women were randomized to treatment AIs as second- and third-line therapy, four comparing an AI to with letrozole versus tamoxifen (12). The exemestane trial was megestrol acetate and two comparing an AI to aminoglutethim- a randomized Phase II trial with a plan to extend to a larger ide. These trials and their results are summarized on Table 1. Phase III trial (13, 14). To date, 120 patients have been ran- More than 2250 women have been randomized to compar- domized to this trial, which is now expanding its accrual in isons of third-generation AIs versus megestrol acetate (4-7). Europe and Canada. Table 2 displays the characteristics of the Superior outcomes of one type or another (OS, TTF, TTP, RR) patients in these trials in terms of ER and PgR positivity, were shown in several of these studies. In addition, >900 numbers with visceral disease, numbers with previous tamox- women were randomized to studies of the second-generation ifen exposure, and numbers having received previous chemo- AIs versus aminoglutethimide (500 mg without prednisone). therapy. The new AIs were shown to be superior for several outcomes in Table 3 displays the major outcome measures from each several of these trials as well (8, 9). study by treatment arm. As demonstrated, RR, CB, and TTP In addition, several Phase III trials comparing third-gener- were superior for the AIs in each study but only significantly so ation AIs head-to-head in postmenopausal women having failed for TTP in the studies comparing anastrozole with tamoxifen tamoxifen are underway. A study conducted by Novartis has and for RR, CB, and TTP in the study comparing letrozole with compared anastrozole with letrozole in >650 women with tamoxifen. The exemestane versus tamoxifen study, although measurable or assessable lesions using TTP, RR, TTF, and OS suggesting large differences in OR and CB, had so few patients as endpoints. Results are not as yet available. In addition, an that significance cannot be determined. Considerable discussion ongoing study conducted by Pharmacia is comparing anastro- has surrounded the differences between the European and the zole to exemestane in women with measurable visceral lesions. American anastrozole versus tamoxifen studies and the differ- More than 200 women have entered this study for which the ences in their patient populations. The North American study stated endpoints are RR, TTP, OS, and tolerability. Once again, suggested a significant difference in TTP and in CB in favor of no results have as yet been published or presented. anastrozole, while the European study did not. Although a much

Table 2 Metastatic breast cancer optimal endocrine Table 4 Metastatic breast cancer anastrozole (A) vs tamoxifen (T) as therapy "upfront" first line hormonal therapy Randomized Phase Ill trials Tolerability data versus tamoxifen Venous Anastrozole Letrozole Exemestane thromboembolic Hot Gastrointestinal No. patients 1021 907 120 (random Study Arms disease flashes disturbances Lethargy (pooled) (1 trial) Phase II) Europe A 5% 21% 23% 1% % patients with T 7% 21% 28% 3% ER or PgR+ tumors 66 66 90 US + A 4% 38% 54% 1% Visceral disease 48 45 56 Canada T 8% 28% 57% 3% Previous tam exposure 20 18 18 Previous CTX exposure 40 32 24

exemestane. In addition, the extensive data showing an 8% RR and a 24% CB for exemestane in patients who have failed Table 3 Metastatic breast cancer optimal endocrine therapy "upfront" previous AIs, including aminoglutethimide and third-generation AIs, are also impressive (15). Such data have, however, not been Randomized Phase III trials versus tamoxifen collected in such a systematic fashion for letrozole or anastrozole, so that the clinical superiority of exemestane in this regard Anastrozole Letrozole Exemestane is really unknown. Clearly more data comparing exemestane Efficacy results 1021 907 120 (random with tamoxifen and comparing the various AIs directly with one (pooling) (1 trial) Phase II) another would be very useful. AI/TAM OR (%) 29/27 30/20" 41/14 In summary, however, it seems that both letrozole and CB (%) 57/52 49/38 '~ 56/42 anastrozole are at least as good and probably better than tamox- TTP 8.5/7.0" 9.4/6.0 ~ N/A ifen for first-line therapy of metastatic disease. For anastrozole, Reference JCO 2000 JCO 2001 ASCO, 2001 perhaps only TTP is superior; however, while for letrozole, a Statistically significant. every major endpoint reported to date shows superiority. Clearly the direct comparison trial between anastrozole and letrozole will provide additional interesting data. Thus, it seems appropriate for women with hormonally higher proportion of the North American patients were known to responsive disease to begin therapy with an AI. Tamoxifen may be ER- and PgR-positive, the patients with ER and PgR status still be acceptable first-line therapy in this setting, however, unknown in the European study were likely quite similar to the particularly if issues of cost or drug availability are paramount. North American patients, so that the lack of measurement of ER There is to date no data suggesting that prescribing such patients and PgR status in some women in the European study does not tamoxifen first and moving to an AI as second line is detrimen- likely explain these small differences. Hopefully, these ER and tal to their overall survival. PgR data can be retrospectively obtained so that this matter can Clearly, additional results from ongoing trials, including be clarified. those in the adjuvant setting, will shed additional light in these Table 4 shows some interesting toxicity data. These data areas. It will also be interesting to follow the upcoming com- suggest that thromboembolic disease is slightly more common parisons of fulvestrant (Faslodex) to tamoxifen since this "pure" in patients randomized to tamoxifen than anastrozole, as perhaps antiestrogen has already been shown in randomized trials to be are gastrointestinal disturbances and lethargy. Hot flashes, how- at least equivalent to anastrozole in patients whose tumors have ever, seem equally common in women in both arms. No tests of already progressed on tamoxifen. significance have been applied to these toxicity differences because of concern about multiple testing. Open Discussion The letrozole versus tamoxifen trial had a somewhat dif- Dr. Kent Osborne: These are relatively small trials, al- ferent design in that a crossover to the other compound was built though they may have 200 or 300 patients in each arm. The into the design of this trial. The primary endpoint was TTP. largest trial is the Femara and tamoxifen trial, and it does show Patients studied appear very similar to those in the anastrozole more consistency. If you look at all those trials, the fulvestrant trial. In particular, the percentages with ER-positive versus versus Arimidex, the Arimidex versus Megace, and all those unknown receptor status for this trial and for the pooled anas- trials, it's always trending. A couple of months could be better. trozole versus tamoxifen trials are virtually identical. The results It's just not statistically significant. If you double the patient of the letrozole versus tamoxifen trial are somewhat more strik- size, it probably would be, and you'd have more consistency. ing in that TTP, TTF, RR, and CB were all highly statistically Dr. Pritchard: Overall, though, if you look at more than significantly better for letrozole. 1000 patients in the two anastrozole trials rolled together, you Survival data are not yet available from either the exemes- don't see consistent changes in the end points; it's probably just tane or the letrozole trial and are awaited with interest. the opposite end of the same confidence interval. I personally The study of exemestane versus tamoxifen, while prelim- remain fairly unconvinced that there are real differences clini- inary, also suggests that there could be a benefit in favor of cally among these third-generation aromatase inhibitors.

Dr. Matthew Ellis: If you look at the number of patients 3. Njar, V. C. O., and Brodie, A. M. H. Comprehensive pharmacology who progress on these trials in 3 months in the second-line and clinical efficacy of aromatase inhibitors. Drugs, 58: 233-255, 2001. setting, it's about 50%. In the first-line setting, it was about 4. Buzdar, A. U., Jonat, W., Howell, A., et al. Anastrozole versus 30%. Metastatic disease is a tough setting in which to develop megestrol acetate in the treatment of postmenopausal women with advanced breast carcinoma: results of a survival update based on a endocrine therapy drugs because there's so much resistance. combined analysis of data from two mature phase III trials. Arimidex There may be very important differences between different Study Group. Cancer (Phila.), 83:1142-1152, 1998. classes of drugs that could be critical in the adjuvant or preven- 5. Kaufmann, M., Bajetta, E., Dirix, L. Y., et al. Exemestane is superior tion setting. You may not see it in the metastatic disease setting to megestrol acetate after tamoxifen failure in postmenopausal women because the disease doesn't care about estrogen at that point. with advanced breast cancer: results of a phase III randomized double- Dr. Pritchard." That's why it is great that we've gone on to blind trial. The Exemestane Study Group. J. Clin. Oncol., 18: 1399- 1411, 2000. do some big studies in the adjuvant setting, and I think the data may be much more reliable. 6. Dombernowsky, P., Smith, I., Falkson, G., et al. Letrozole, a new oral aromatase inhibitor for advanced breast cancer: double-blind ran- Dr. Per Lonning: I agree with you that very many patients domized trial showing a dose effect and improved efficacy and tolera- don't respond to any type of therapy, particularly when they bility compared with megestrol acetate. J. Clin. Oncol., 16: 453-461, come to the second-line treatment, but in general you see im- 1998. proved duration of the response with all these novel compounds. 7. Buzdar, A., Douma, J., Davidson, N., et al. Phase III, multicenter I think all of us believe that one of the key reasons why these double-blind, randomized study of letrozole, an aromatase inhibitor, for new drugs have shown superiority is that they are more potent advanced breast cancer versus megestrol acetate. J. Clin. Oncol., 19: 3357-3366, 2001. biochemically. With the comparison of anastrozole and letro- 8. Gershanovich, M., Chaudri, H. A., Campos, D., et al. Letrozole, a zole, the question will be how that fits with the biochemical new oral aromatase inhibitor: randomized trial comparing 2.5 mg daily, data. In our head-to-head comparison, letrozole is much more 0.5 mg daily and aminoglutethimide in postmenopausal women with potent compared to anastrozole, both with respect to aromatase advanced breast cancer. Letrozole International Trial Group (AR/BC3). inhibition and estrone sulfate suppression. Another interesting Ann. Oncol., 9: 639-645, 1998. question is the lack of cross-resistance between the compounds, 9. Bengtsson, N. O., Focan, C., Gudgeon, A., et al. A phase III trial because there are five different trials to show the lack of cross- comparing vorozole (Rivizor) versus aminoglutethimide in the treatment of advanced postmenopausal breast cancer. Eur. J. Cancer, 33: S148, resistance between a steroid and a nonsteroidal. Contrary to 1997. common belief, it goes both ways, and this lack of cross resist- 10. Nabholtz, J. M., Pollak, M., Harwin, W., et al. Anastrozole is ance even obtains if you use a less potent compound after the superior to tamoxifen in first-line therapy for advanced breast cancer in more potent one up front. So it relates to difference in the postmenopausal women: results of a North American multicenter ran- mechanism of action of the drugs. domized trial. J. Clin. Oncol., 18: 3758-3767, 2000. Dr. Anthony Howell: If I might make two comments, one 11. Bonneterre, J., Thurlimann, B., Robertson, J. F., et al. Anastrozole is that the mistake is to do two trials. When you do one on one versus tamoxifen as first-line therapy for advanced breast cancer in 668 side of the Atlantic and one on the other side of the Atlantic, you postmenopausal women: results of the Tamoxifen or Arimidex Ran- domized Group Efficacy and Tolerability study. J. Clin. Oncol., 18: always get different results. For example, letrozole was highly 3748-3757, 2000. effective compared to megestrol acetate where I come from, 12. Mouridsen, H., Gershanovich, M., Sun, Y., et al. Superior efficacy whereas in the States there was no difference between megestrol of letrozole versus tamoxifen as first-line therapy for postmenopausal acetate and letrozole. But the second issue is the Fuqua mutation women with advanced breast cancer: results of a phase III study of the and how it might relate to responsiveness, which potentially can International Letrozole Breast Cancer Group. J. Clin. Oncol., 19: 2596- change the whole way we look at aromatase inhibitors. It may be 2606, 2001. that those extra picomoles, as Kent Osborne said earlier today, 13. Paridaens, R., Dirix, L. Y., Beex, L., et al. Exemestane (Aromasin) are tremendously important. is active and well tolerated as first-line hormonal therapy (HT) of metastatic breast cancer (MBC) patients (Pts): results of a randomized phase II trial. Proc. Am. Soc. Clin. Oncol., 19: 316, 2000. References 14. Lohrisch, C., Paridaens, J., Dirix, L. Y., et at. No adverse impact on 1. Jensen, E. V., and Jacobson, H. I. Fate of steroid estrogens in target serum lipids of the irreversible aromatase inactivator Aromasin (ex- tissues. In: G. Pincus, and E. P. Vollmer (eds), Biological Activities of emestane) in 1st line treatment of metastatic breast cancer: companion Steroids in Relation to Cancer, pp. 161-178. New York: Academic study to a European Organization of Research and Treatment of Cancer Press, 1960. Breast Group trial. Proc. Am. Soc. Clin. Oncol., 20: 43a, 2001. 2. Osborne, C. K., Yochmowitz, M. G., Knight, W. A., et al. The value 15. Lonning, P. E., Bajetta, E., Murray, R., et al. Activity of exemestane of estrogen and progesterone receptors in the treatment of breast cancer. in metastatic breast cancer after failure of nonsteroidal aromatase inhib- Cancer (Phila.), 46: 2884-2888, 1980. itors: a phase II trial. J. Clin. Oncol., 18: 2234-2244, 2000.

Downloaded from clincancerres.aacrjournals.org on September 25, 2021. © 2001 American Association for Cancer Research. The Role of Tamoxifen and Aromatase Inhibitors/Inactivators in Postmenopausal Patients

Kathleen I. Pritchard

Clin Cancer Res 2001;7:4356s-4359s.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/12/4356s

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/7/12/4356s. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.