Cancer Chemother Pharmacol. 2008 Sep;62(4):667-72. Epub 2007 Dec 7.
Dose-Dense and/or Metronomic Schedules of Specific Chemotherapies
Consolidate the Chemosensitivity of Triple-Negative Breast Cancer: A
Step Toward Reversing Triple-Negative Paradox
Rita S. Mehta
Departments of Division of Hematology/Oncology, Department of Medicine,
Chao Family Comprehensive Cancer Center, University of California at
Irvine School of Medicine, Irvine, CA
To the Editor:
Liedtke et al1 report a progression-free survival of 63% in patients with triple-negative breast cancer predominantly treated with anthracyline-based first and second regimens in the neoadjuvant setting. They suggest that third generation regimen are important considerations in triple-negative subtypes of breast cancer in their discussion, but we feel it needs to be emphasized that third generation regimens should be standard in treatment of triple-negative breast cancer. Combining the four subgroups (based additionally on progesterone receptor status and human epidermal growth factor receptor 2 [HER-2] status) of 483 patients with estrogen receptor–negative breast cancer, the 3-year progression-free survival is similarly 63% and overall survival is 74%.1 These combined group results are comparable with the 5-year progression-free survival of 63% and overall survival of 68% in 327 patients with estrogen receptor–negative breast cancer updated by Citron et al2 in the adjuvant setting (possibly smaller tumors, but longer time outcome) using the standard once-every-3-weeks doxorubicin, cyclophosphamide, and paclitaxel. In the comparator arm, the 5-year progression-free survival improved statistically significantly to 70% and overall survival to 75% in 336 patients with estrogen receptor–negative (includes triple-negative subset) breast cancer with use of dose-dense (once every 2 weeks) administration of doxorubicin, cyclophosphamide, and paclitaxel—a third generation regimen (23% hazard reduction).2 Another trial showed an estimated 5-year event-free survival of 71% in the triple-negative cohort treated with two cycles of dose-dense epirubicin and cyclophosphamide followed by high-dose cyclophosphamide, thiotepa, and epirubicin chemotherapy compared with only 26% in the four cycles of dose-dense epirubicin and cyclophosphamide followed by dose-dense cyclophosphamide, methotrexate, and flurouracil arm, but the poor survival in comparator arm suggest that rapid cycling to an alternate effective regimen may be the underlying mechanism of superiority of dose-dense and dose-intensified regimen.3 Similarly, weekly paclitaxel significantly improved the 5-year progression-free hazard rate by 40% in hormone receptor–negative (including triple-negative subset) breast cancer and 20% in hormone receptor–positive subsets compared with these subsets treated with once-every-3-weeks scheduling.4 A parallel phase III neoadjuvant study demonstrated a pathologic complete response of 43% with denser doxorubicin (weekly) and cyclophosphamide (continuous) compared with 26% with once-every-3-weeks doxorubicin and cyclophosphamide, both arms receiving weekly paclitaxel in locally advanced breast cancer, suggesting an additive benefit of denser administration doxorubicin and cyclophosphamide, in addition to the denser administration of paclitaxel.5 Taken together, studies show benefit of accelerated schedules (weekly or once every 2 weeks) of doxorubicin, cyclophosphamide, and paclitaxel. Importantly, comparing across trials, weekly paclitaxel achieves a higher hazard rate reduction than once-every-2-weeks paclitaxel when both accelerated schedules are compared to once-every-3-weeks paclitaxel.2,4
Of note, Liedtke et al1 show that 22% of patients who achieved pathologic complete response had an overall survival of 94%, and 78% of patients who did not achieve a pathologic complete response had a 68% overall survival—a 26% difference. An absolute 50% increment in pathologic complete response over a baseline of 22% would move an additional 50% of patients from 68% to 94% survival (an absolute 26% improvement for the 50% patients), a 13% increment in survival above a baseline 3-year survival of 74% for 100 patients would result in overall survival improvement to 87% (a 25% increment in pathologic complete response would move 25% of patients from 68% to 94% survival, netting a 6.5% improvement in 3-year survival; ie, each 10% increment in pathologic complete response will translate into an absolute 2.6% 3-year survival). Green et al6 show doubling of pathologic complete response from 23% to 48% with weekly paclitaxel, and an absolute 25% improvement of pathologic complete response in the hormone receptor–negative subset. On projecting, this will translate into an absolute 6.5% survival benefit, that is an improved 3-year overall survival of 81.5%, a 25% hazard reduction at a minimum. Speculatively, additional survival benefit may accrue with increase in the minimal residual disease.6 Indeed, the actual benefit in the trial reported by Sparano et al 4 was 40% hazard reduction with weekly paclitaxel compared to once-every-3-weeks paclitaxel in hormone receptor–negative subset.
Of the prospectively maintained database of 14 patients with stage IIA-IIIB (including inflammatory breast cancer) or local relapse HER-2–negative, estrogen receptor–negative, and progesterone receptor–negative, infiltrating ductal carcinoma (excluding two patients with metaplastic carcinoma) of breast who received dose-dense doxorubicin and cyclophosphamide (maximum four doses, except one who relapsed after doxorubicin, cyclophosphamide, and flurouracil) followed either by dose-dense paclitaxel (n = 2; four doses), or by weekly paclitaxel (cremophor [80 mg/m2] or albumin bound [100 mg/m2]) with carboplatin (at area under the curve of 2, both 3 weeks on, 1 week off; maximum 12 doses, n = 12) plus or minus maximum eight doses of bevacizumab every 2 weeks (n = 4; maximum 8 doses), 10 of 14 patients (71%; 95% CI, 42% to 92%) achieved pathologic complete response. Two additional patients had scattered cells in the breast with no lymph node involvement. Thus, 13 of 14 (93%; 95% CI, 66% to 100%) patients had no evidence of residual lymph node involvement. The progression-free survival of 86% (95% CI, 57% to 98%) and overall survival of 93% (95% CI, 66% to 100%) of 14 patients at a median follow-up of 36 months (range, 22 to 68 months) compared with the projected survival improvement of 87% based on pathologic complete response of 71% is promising. Assessed another way, a reversal of pathologic response incidence from 22% reported by Liedtke et al1 to 71% is likely to show improvement in survival closer to 94% for the majority of patients, a result seen with our data—a 73% hazard reduction. Therefore, the high pathologic complete response reported by us parallels the benefit of once-every-2-weeks anthracyclines (or denser), and once-every-2-weeks or once-weekly paclitaxel scheduling in phase III trials.5,6 Faster delivery of anthracyclines allows administration of targeted or alternate effective treatment earlier. For example, hormone receptor blocker for hormone receptor–positive breast cancer, trastuzumab for HER-2–positive breast cancer, and once-weekly paclitaxel, an alternate effective treatment for all subsets of breast cancer, specifically triple-negative subset. Therefore, for patients with triple-negative tumors, optimally scheduled doxorubicin and cyclophosphamide (once every 2 weeks) and paclitaxel (once weekly) should be standard.2,4 As 12 of 14 patients also received carboplatin, the use of carboplatin may be considered in any patient with large (as seen in neoadjuvant setting) triple-negative breast cancer.7,8 While carboplatin based adjuvant phase III studies are limited to docetaxel, trastuzumab, and carboplatin combination, extrapolation of docetaxel, trastuzumab, and carboplatin combination to paclitaxel and carboplatin (and trastuzumab in HER-2–positive breast cancer) after anthracyclines may be reasonable pending large adjuvant or neoadjuvant trials in triple-negative breast cancer. Importantly, carboplatin is standard treatment for ovarian and lung cancer, and therefore long-term outcome is known. Moreover, the BRCA-ness of the triple-negative tumors makes these tumors specifically susceptible to the DNA-damaging action of carboplatin give support to our use of carboplatin in triple-negative breast cancer.9 Additional small studies that used platinums have similarly shown a promising outcome either in pathologic complete response (20% with single-agent cisplatin to 67% with platinum combination regimens) or in progression-free survival (75% to 84%) in triple-negative breast cancer.10-12,14-15 We did not see any progression while patients were receiving chemotherapy, unlike the studies by Torissi et al10,13 where they show a substantial progression while patients were either on weekly paclitaxel or dose-dense docetaxel regimen. This may be secondary to delayed switch to alternate effective therapy of weekly paclitaxel in the first case, or switch to a less effective dose-dense docetaxel switch in the second case.
The potential additional benefit of bevacizumab over optimum scheduling of standard chemotherapy needs to be tested in randomized trials, as the follow-up is shortest for the four patients receiving bevacizumab. While the small number, short follow-up, and heterogeneity of chemotherapy are limitations of our small study, the chemotherapy was uniformly delivered in accelerated fashion. Achievement of high pathologic complete response and the corresponding survival benefit tied to high pathologic complete response are the strengths of the study. Despite a short follow-up, the follow-up would be considered adequate for progression-free survival in triple-negative subtype of breast cancer as the hazard rates of relapse are highest in the first year, and taper off quickly after 3 years.1 We emphasize that third generation chemotherapeutic regimens are the most important treatment regimens for triple-negative breast cancer in achieving a higher pathologic complete response, and thereby higher survival, where the only current targeted treatment is the accelerated and optimized chemotherapy that targets the Gompertzian growth model as first shown by Norton et al.16 Moreover, poor clinical outcome seen in patients with triple-negative despite initial chemosensitivity as demonstrated by Carey et al17 can be overcome by consolidating the gains obtained by initial chemosensitiviy of accelerated anthracyclines with accelerated optimal chemotherapy, which at a minimum should include weekly paclitaxel.