Poster Discussion Section: Systemic Management

Tari King, MD

Tiffany A. Traina, MD

Conleth Murphy, MD

Post by: Tari King, MD; Tiffany A. Traina, MD, Conleth Murphy, MD
Post from: ASCO Breast Cancer Symposium

Prior to the Symposium there was very little data concerning the issue of whether or not women with small, lymph node negative, HER2-positive breast cancers derive benefit from adjuvant trastuzumab. McArthur and colleagues from MSKCC reported results of a retrospective study comparing the management and outcomes of HER2+, T1N0 tumors in the trastuzumab and pre-trastuzumab era (abstr #228).  Of the 256 cases reviewed, more than one-third of patients had subcentimeter tumors. Baseline characteristics between cohorts were similar with the exception of chemotherapy use, which was greater in the trastuzumab-era patients. After 4 years of followup, patients with small, node-negative, HER2+ breast cancers had excellent outcomes following adjuvant chemotherapy with trastuzumab with 3 year DFS 100% and OS 99%. This strongly suggests a major impact of trastuzumab therapy in these patients.

Safety and efficacy after 6 years of followup were reported for the neoadjuvant, randomized trial comparing paclitaxel followed by FEC, with or without concurrent trastuzumab in patients with HER2+ breast cancer (abstr #221). No clinical cardiac toxicity events have occurred in patients receiving anthracycline and trastuzumab during this long-term followup period. The overall survival advantage to trastuzumab in the setting of anthracycline/taxane chemotherapy was maintained at 6 years (84% vs. 100%, p=0.035).

Metastatic Breast Cancer Oral Session

Post by Chau T. Dang, MD:

Robert et al reported the results of RIBBON-1 in which over 1200 patients with metastatic breast cancer were randomized to treatment with chemotherapy (CRx) with the anti-angiogenic agent, bevacicumab (bev) or placebo in a 2:1 ratio in the front-line setting. Chemotherapy included a range of options such as capecitabine (Cape), a taxane (T, nab-paclitaxel, docetaxel), or an anthracycline (A). The primary endpoint was progression-free survival (PFS). The study demonstrated an improvement in PFS when Bev was added to chemotherapy and this was consistent with prior studies of similar design. At a median FU of 15.6 months in the Capecitabine cohort and 19.2 months in the T+A cohort, there was an improvement in PFS when Bev was added to chemotherapy. PFS was 5.7 vs 8.6 months (p 0.0002) for Capecitabine + Bev and was 8.0 vs 9.2 months (p<0.0001) for T/A + Bev. The RRs were better in Cape + Bev and T/A + Bev vs the respective chemotherapy alone, and this was statistically significant. Again there was no impact on overall survival.

Andy Seidman (from our group) and investigators around the U.S. reported the results of a phase 3 randomized trial of docetaxel plus capecitabine vs docetaxel plus gemcitabine. Uniquely in this study, when their cancer worsened patients could cross-over to the other single agent chemotherapy drug. The study suggests that the specific sequence of agents used in the metastatic setting may not be critical in terms of overall outcomes for patients and provides support for a range of individually tailored approaches.

Also from our team, Dr. Alison Conlin (a recent graduate of our fellowship training program) reported the results of a randomized phase II study which tested Bev + nab-pacliaxel (nab-P) at 3 different schedules (Arm A: administered the taxane at standard doses every third week, Arm B was identical except the dosing was every other week and supported by growth factor (filgrastim) while Arm C was half dose given every week. A total of 209 patients were accrued and while Arm B was closed early for toxicity, there were no significant differences in efficacy. We concluded that with the possibly longer TTP seen with the weekly schedule of nab-P + Bev (not statistically significant however) coupled with the lesser toxicity of this arm, makes it the preferred regimen for future studies, perhaps using an intermittent, 3 on/1 off schedule. This weekly nab-P + Bev regimen is one arm of an on-going phase 3 trial of weekly chemotherapy plus Bev (CALGB 40502).

PARP inhibitors and Loco-regional/Adjuvant Therapy

Post by Maura N. Dickler, MD:

A major story was the introduction of a new class of drug for the treatment of breast cancer. These drugs are called “PARP inhibitors” and they prevent cells from repairing a specific kind of genetic damage. Laboratory experiments suggested that breast cancers arising in patients with inherited mutations in the BRCA genes would be particularly sensitive and one trial (conducted in part at MSKCC) testing a PARP inhibitor (Olapaprib) in such patients demonstrated impressive activity as predicted. At the plenary session, Joyce O’Shaughnessy presented results from a randomized trial of gemcitabine and carboplatin with or without another PARP inhibitor called BSI-021. This trial went beyond patients with BRCA mutations and instead enrolled patients with triple negative metastatic breast cancer, as defects in DNA repair are thought to play a role in this breast cancer subtype. Among the 123 patients enrolled, the addition of the PARPi increased the response rate from 16% to 48% (p=0.002) and increased the clinical benefit rate from 21% to 62% (p=0.0002); median PFS was increased from 3.3 to 6.9 months and overall survival from 5.7 to 9.2 months (both significant) in favor of the PARPi arm as well with no clear increases in toxicity. A phase 3 trial is planned for this summer.

On Sunday morning, the poster discussion session was focused on loco-regional and adjuvant therapy. Desmedt et al reported the results of the neoadjuvant TOP trial which is a prospective trial designed to identify markers of response and resistance to preoperative epirubicin at 100 mg/m2 q 2 or 3 weeks in patients with ER(-) breast cancer. Patients had prechemotherapy breast biopsies, and TOP2A was analyzed by FISH (Abbott triple probe) and IHC (KiS1) and gene expression profiles were generated (HG133Plus2.0). In this study 149 patients received neoadjuvant epirunicin. In this study 32 percent of the patients had HER2 amplified tumors. The TOP2A amplification (11 percent of the cases) was seen exclusively in HER2 cases was highly predictive of pCR. Overall, half of the patients with TOP2A amplification had a pCR (p 0.001), but the protein and mRNA levels were not predictive. Chromosome polysomy 17 (Ch 17) was seen in 68 percent of the patients and was not predictive a pCR. The TOP2A gene expression module made of genes located closely to TOP2A was predictive of response in HER2+ patients in this trial ( p 0.00001). To validate their result, the authors also generated the gene expression profile from another trial (BIG-00) which compared a preoperative anthracycline (A) to a taxane (T). They found that the TOP2A gene expression module was also predictive of response in the ER-/HER2+ patients in the BIG-00 trial treated with an A, but not with a T ( p 0.0005). Stroma expression profile was also predictive of response in both studies. This trial shows that TOP2A gene amplification or amplification of genes proximate to TOP2A may be a predictive marker of response to an anthracycline. Further work will be needed to confirm this result and if validated, perhaps TOP2A amplification may be used in the future to help tailor treatments for our patients. This is particularly relevant to work done at MSKCC and Cold Spring Harbor first reported last December at San Antonio in which we are able to examine these genes with greater precision using representational oligonucleotide microarray (ROMA) technology.

Breast Ca: loco-regional disease and adjuvant therapy

Post by Tiffany A. Traina, MD:

Aubert et al performed a retrospective review of pharmacy and medical claims to describe the impact of drugs that inhibit CYP2D6 (commonly used antidepressants) given concurrently with tamoxifen. Because the enzyme CYP2D6 converts tamoxifen into an active form, inhibitors of this enzyme theoretically could reduce the benefits of taking tamoxifen. These investigators found an increase in breast cancer recurrence with use of moderate to potent CYP2D6 inhibitors (including some commonly used anti-depressants). A small sample size and limited follow-up make this report less than definitive. In addition, CYP2D genotype (normal variations in CYP2D6 function not related to specific drugs) was not available for patients in this cohort. A second study failed to confirm any impact on outcomes. Hence, there are no definitive data that conclusively links the use of CYP2D6 inhibitors to reduced benefit from endocrine therapy. Nevertheless, because of the theoretical possibility that CYP2D6 inhibitors could possibly reduce tamoxifen efficacy, it is prudent to avoid concurrent use whenever possible, as we are already doing at MSKCC

The effect of endocrine therapy on cognitive function in postmenopausal women with early stage breast cancer was reported by Ribi et al. After 5 years of study therapy on BIG1-98, a subset of patients randomized to receive tamoxifen (T), letrozole (L) or these agents in sequence (T–L or L–T) underwent a series of cognitive function tests. Data was analyzed to compare results of patients who received L vs. T during the past 3 years. Investigators found that fewer patients had impaired cognitive function when taking letrozole compared with tamoxifen (37% v 54%, p=0.05). Limitations of the study include the lack of baseline cognitive function testing, possible differences in chemotherapy use, the small numbers of patients involved and the exclusion of patients who had withdrawn from the study.

A 70-gene risk profile was applied to fresh-frozen tissue from 1,600 early-stage breast cancer patients in a large meta-analysis of 7 adjuvant trials of endocrine therapy with or without chemotherapy. This assay categorizes patients into low-risk or high-risk of recurrence. In the 541 patients with 0-3 involved LNs, Bender et al report that chemotherapy plus endocrine therapy lowered the risk of recurrence of breast cancer elsewhere in the body for those patients categorized as high-risk but did not add to endocrine therapy for low-risk patients. The problem with this data, discussed by several questioners, was the fact that the assignment to chemotherapy or not was not done randomly. Hence other factors, not accounted for, might explain the differential effect of chemotherapy observed here. The MINDACT study will explore the use of this risk signature profile prospectively to determine use of adjuvant chemotherapy in addition to endocrine therapy.

Esserman et al reported on the I-SPY study, which is a multicenter trial (offered at MSKCC) designed to identify markers predictive of pCR and survival for women w/ LABC. Women were given neoadjuvant AC followed by paclitaxel. The early endpoints include tumor response on MRI, pathologically documented complete response, and residual cancer burden (RCB) and late endpoints include freedom from recurrence and survival. Overall, 213 patients were enrolled to have serial breast MRIs and core biopsies, and 149 were analyzed to date. Pretreatment assays included Agilient expression assays, MIP aCGH, p53 gene chip and sequencing, and IHC and reverse phase protein arrays (RPMA). Response to therapy was assessed by breast MRI, pCR, and RCB. Most pts had poor prognostic signatures using the 70-gene risk profiled discussed above. As predicted, patients with triple negative breast cancer had a high pCR and those with good risk features (ie: luminal A) had a low response. Patients with basal-like tumors who had an excellent response had better outcomes than those who did not. PCR and RCB were highly predictive of outcomes. I-SPY reported that many of these biomarkers predicted response to therapy and outcomes. Additional analyses on this study will be reported tomorrow.

Metastatic Breast Cancer

Post by Shanu Modi, MD

Breast cancer abstracts were presented at a poster discussion session focused on the characterization and treatment of patients with metastatic breast cancer.

Rugo et al. reported a randomized multicenter phase 2 study of ixabepilone, an epothilone B analog, in combination with bevacizumab (ixa, weekly and q 3 weeks). Both of the ixabepilone/bev arms were compared to weekly paclitaxel/bevacizumab (control arm), and were found to have similar antitumor activity and safety in patients receiving first-line chemotherapy for metastatic disease. This data supports the feasibility of an ongoing phase 3 trial within the CALGB (40502), which is powered to differentiate the efficacy between these regimens.

Gluck et al performed an exploratory analysis of outcomes for patients receiving docetaxel and capecitabine as compared to docetaxel alone based on their estrogen receptor (ER)status. The addition of capecitabine to docetaxel improved OS in patients with ER+ tumors but not in those that were ER negative. This adds to the existing literature of unplanned retrospective subset analyses suggesting increased capecitabine efficacy for patients with ER+ tumors. We caution against drawing any conclusions about capecitabine activity based on hormone receptor status due to the retrospective nature of these data.Palacova and colleagues found patients with triple negative tumors to be at higher risk for early development of CNS disease. This raises questions regarding the need for predictors of patient subsets at risk of CNS disease and the role for CNS screening and interventions.

Two important studies evaluating novel HER2 targeting agents were presented. Vogel et al reported the final results from their phase 2 study of the antibody-conjugate T-DM1for patients with pre-treated HER2+ MBC. This compound first showed promise in a phase 1 trial and this activity is now confirmed in phase 2. 112 patients with HER2+ MBC, receiving a median duration of 18 months of prior trastuzumab therapy, 60% pre-treated with lapatinib, and a median of 3 prior cytotoxic agents were treated with single agent T-DM1. An independent review confirmed a response rate of 25% and clinical benefit rate of 35%. Equally important is that this treatment is well tolerated with reversible thrombocytopenia (low platelet counts) as its predominant toxicity. Phase 3 trials are underway already and we look forward to further development of this promising new agent.

A second HER2 targeting agent, pertuzumab, is a monoclonal antibody which inhibits HER2 dimerization and has previously shown strong activity in combination with trastuzumab (RR 24%). Cortes and colleagues presented single agent efficacy data which showed only a modest 7% overall response rate for pertuzumab in pre-treated patients with HER2+ metastatic disease. We should expect further studies and development of this agent in combination with trastuzumab moving forward.