Findings Support Nab-Paclitaxel as Top Choice for First-Line TNBC Therapy

Hope Rugo, MD

As the search for new therapies for patients with recurrent triple-negative breast cancer (TNBC) continues, recent research demonstrates that better outcomes may be achieved through changes in existing chemotherapy protocols.

In long-term findings of a phase III study, the median overall survival (OS) for patients with TNBC was 21.0 months with nab-paclitaxel (Abraxane) compared with 15.3 months with standard paclitaxel, representing a 26% reduction in the risk of death, with a hazard ratio of 0.74 (95% CI, 0.51-1.07). The median progression-free survival (PFS) was 7.4 months with nab-paclitaxel and 6.4 months with paclitaxel, which translated into a 21% reduction in the risk of progression (95% CI, 0.55-1.12).¹

The results were reported for a subgroup of the CALGB 40502/NCCTG N063H trial, which enrolled 799 patients with locally recurrent or metastatic stage IIIC or IV breast cancer.

Participants were randomized to receive paclitaxel (n = 283), nab-paclitaxel (n = 271), or ixabepilone (Ixempra) (n = 245) as first-line therapy in this setting. Most patients on the trial also received concurrent bevacizumab (Avastin), which at the time was the standard of care in the first-line setting. Nab-paclitaxel was given at a higher-than-approved dose of 150 mg/m² weekly. Paclitaxel was administered at 90 mg/m² and ixabepilone was administered at 16 mg/m2 weekly.

OncLive: Can you provide an overview of the design of the study?

The efficacy findings for the chemotherapy regimens varied by subtype. “For hormone receptor—positive disease, paclitaxel remains the gold standard. For TNBC, which is more resistant, I would consider nab-paclitaxel as a first option in the first-line metastatic setting...,” said lead investigator Hope S. Rugo, MD, medical oncologist and director of the Breast Oncology Clinical Trials Program at the University of California, San Francisco, Helen Diller Family Comprehensive Cancer Center. In an interview with OncLive^® at the 2017 San Antonio Breast Cancer Symposium (SABCS), Rugo discussed the implications of these findings and the importance of individualizing doses of nab-paclitaxel.Rugo) This is a trial that we started quite a long time ago, and it was designed to evaluate the relative efficacy of novel microtubule inhibitors given in a weekly schedule, compared with the standard of care, which was paclitaxel. At the time, bevacizumab was the standard of care as first-line therapy for metastatic HER2-negative breast cancer, based on the results of ECOG 2100. Our control arm was weekly paclitaxel at 90 mg/m2 with bevacizumab, 3 weeks on, 1 week off. If patients had responding or stable disease after 6 cycles of 4 weeks per cycle, they could stop the chemotherapy and continue on with bevacizumab alone.

One of our experimental arms was weekly ixabepilone, which had been approved based on efficacy in taxane-resistant disease, as well as superiority in combination with capecitabine (Xeloda). However, it was given every 3 weeks and was quite toxic. We were evaluating weekly therapy at the same dose—16 mg/m². The second experimental arm was nab-paclitaxel. A randomized phase II trial conducted in Russia and published by William Gradishar, MD, had shown improved efficacy with 150 mg/m², compared with paclitaxel, every 3-week paclitaxel, and docetaxel (Taxotere). The idea was that we would study that dose. In our study, each experimental arm was compared with the control, rather than to each other.

What were the results of this study?

Was a difference seen in the subset analysis?

We had planned to enroll 900 patients with a primary endpoint of PFS, but we had interim analyses for futility. At the first interim analysis, the ixabepilone arm was closed for futility, and at the second interim analysis, the entire study was closed because of futility.As one would expect, the first time we presented our data at SABCS in 2013, and when they were published in the Journal of Clinical Oncology in 2015, we showed that neither arm was superior to paclitaxel, and that the nab-paclitaxel arm, in particular, was associated with increased hematologic and nonhematologic toxicity, including both motor and sensory neuropathy. We don’t usually see much motor neuropathy. We also found that the ixabepilone arm was inferior for PFS, and that neither arm showed any difference in OS.We also looked at the subsets based on hormone receptor [HR] status—HR-positive, HER2-negative, or TNBC. What we found was a significant test for interaction between nab-paclitaxel and paclitaxel for HR status, and no significant interaction for ixabepilone for both PFS and OS.

When we looked at the specifics of the subset analysis, we found a trend toward superiority of nab-paclitaxel compared with paclitaxel in the TNBC group only. In the HR-positive group, which was a larger group where we have more power to detect significance, we found that both nab-paclitaxel and ixabepilone trended toward being inferior to the paclitaxel arm. The same was true for OS, where ixabepilone was inferior and nab-paclitaxel was trending toward inferiority for OS in the HR-positive subset but was numerically quite superior for OS in the TNBC group, at 21 versus 15 months, which was quite intriguing. We continued to show that there were more dose reductions and dose discontinuations for the nab-paclitaxel arm, which was concordant with the increased toxicity.

This was all in light of the GeparSepto data, which showed that nab-paclitaxel, which they dose-reduced from 150 mg/m² to 125 mg/m², was superior for event-free survival in both TNBC and HR-positive disease. In terms of PFS, we saw a trend toward a differential benefit in TNBC versus HR-positive disease. It may be that in disease that has previously seen a taxane and is highly resistant, like TNBC, there is some advantage to nab-paclitaxel, or that there was an advantage of dose intensity.

I think the take-home message from this is that in certain settings, there might be an advantage for nab-paclitaxel versus paclitaxel, but we don’t know exactly what they are. For HR-positive disease, paclitaxel remains the gold standard. For TNBC, which is more resistant, I would consider nab-paclitaxel as a first option in the first-line metastatic setting, keeping in mind that 150 mg/m² should never be the dose, because it is associated with too much toxicity. The highest dose that should be considered is 125 mg. Even at that dose, there was an 8% rate of peripheral neuropathy in the grade 3 range in the GeparSepto study, which is always a difficult toxicity for the patients. As always, we need to balance toxicity versus benefit. At the moment, we don’t have any specific setting where it would be incorrect to give paclitaxel as your first treatment option.