Kevin Kalinsky, MD, MS

Dawn Hershman, MD, MS

THE IMPASSION130 trial—reported in The New England Journal of Medicine by Schmid et al¹ and reviewed in this issue of The ASCO Post—was an eagerly awaited study in newly diagnosed metastatic triple-negative breast cancer. To briefly review, 902 patients were randomly assigned in a 1:1 fashion to receive nab-paclitaxel at 100 mg/m2 (3 weeks on/1 week off) plus atezolizumab at 840 mg (every other week) vs nab-paclitaxel plus placebo. Although there was an overall statistically significant improvement in median progression-free survival with the addition of atezolizumab, with an absolute median benefit of 1.7 months, the benefit was restricted to the programmed cell death ligand 1 (PD-L1)–positive cohort, with an absolute median benefit of 2.5 months and a 1-year improvement of 12.7%. Notably, in this subset analysis, there was a nearly 10-month improvement in overall survival with the combination in the PD-L1–positive group.

Some may ask, “Why so much enthusiasm over such a small absolute benefit?” This study offers definitive evidence that there is efficacy of immunotherapy for a subset of patients with triple-negative breast cancer, a tumor subtype for which there has been an unmet need to identify therapeutic options beyond chemotherapy. Prior to this study, the limited benefit in patients with pretreated metastatic breast cancer had left many clinicians uncertain whether this approach would ultimately be an effective one for patients with breast cancer. We have learned that the earlier in the disease course patients with breast cancer are treated with immunotherapeutic strategies, the more likely a response will be observed. Given the findings of the IMpassion130 trial, the U.S. Food and Drug Administration granted a Priority Review designation for this combination for front-line metastatic triple-negative PD-L1–positive breast tumors.

“We have learned that the earlier in the disease course patients with breast cancer are treated with immunotherapeutic strategies, the more likely a response will be observed.”

— Kevin Kalinsky, MD, MS, and Dawn Hershman, MD, MS

Tweet this quote

Biomarkers of Response

ALTHOUGH THE PD-L1 subset analysis was preplanned and hypothesis-driven, it should be noted that the reported survival analysis was not presented with a formal statistical test and P value. It is important to note that a tumor was deemed as “positive” for PD-L1 based on the SP142 immunohistochemical assay by Ventana Medical System (≥ 1% expression). The PD-L1–positive group made up 41% of the total study population and was equally distributed between the two arms.

At the 2018 San Antonio Breast Cancer Symposium (SABCS), it was reported that the prevalence of PD-L1 was expressed more on immune cells than tumor cells, and this expression was associated with response to therapy.² Of the total population, only 14% had PD-L1 staining of at least 5% on tumor-infiltrating immune cells. The hazard ratio for benefit with the addition of atezolizumab was similar (~0.60) between those with PD-L1 of at least 1% but less than 5% and those with at least 5%—although these sample sizes were relatively small. Future research will continue to focus on additional markers that may help further discriminate the likelihood of response to checkpoint inhibition when combined with chemotherapy.

At the 2018 SABCS, other subgroups of biomarkers were evaluated.² Expression of CD8 was enriched in PD-L1 positive tumors, and patients with CD8-positive tumors achieved clinical improvement with the addition of atezolizumab only if the tumor co-expressed PD-L1. The same held true for stromal tumor–infiltrating lymphocyte expression: patients with tumor-infiltrating lymphocyte–positive tumors derived benefit only in PD-L1–positive tumors. Lastly, patients with BRCA1/2 mutations responded only if their tumors were PD-L1–positive. Thus, PD-L1 immune cell expression remains the best predictor of clinical benefit at this time.

Immune-Related Toxicity

ALTHOUGH UNCOMMON, immune-related adverse events can occur and should be carefully explained to patients. These side effects include a 10% higher rate of hypothyroidism and an 8% higher rate of pyrexia compared with the control arm. Infrequent signs or symptoms of other immune-related side effects, such as pneumonitis and hepatic dysfunction, were also found.

“PD-L1 immune cell expression remains the best predictor of clinical benefit at this time."

— Kevin Kalinsky, MD, MS, and Dawn Hershman, MD, MS

Tweet this quote

With the increased utilization of checkpoint inhibitors in other solid tumors, it has become clear that when administered to the general population (as opposed to those carefully selected to enroll in a trial), the incidence and severity of side effects can be higher than those reported in clinical trials. Careful attention should be paid to the signs and symptoms of potentially life-threatening toxicities. In addition, the potential of financial toxicity should be explained to patients who are considering this costly medication while weighing the risks and benefits of receiving this combination.

Questions Remain

A NUMBER of questions remain. Unquestionably, there is a subgroup of patients who respond to the addition of atezolizumab. What will be the durability of that response? Will some patients have long-lasting responses? Can any other markers beyond PD-L1 assist in predicting these durable responses? For patients with a durable response, is there a point in time that they can stop the checkpoint inhibitor? In addition, will nab-paclitaxel remain the best cytotoxic partner with checkpoint inhibition? The theoretical benefit of nab-paclitaxel is that patients do not need to receive steroids with its administration, potentially impacting T-cell function. However, in this population, there will be patients whose disease progresses quickly after receiving a taxane in the operable setting. Perhaps, a different chemotherapy partner may offer the same (or more) benefit—including one not associated with worsening neuropathy? Also, will there be a future role for immunotherapy in breast cancer beyond triple-negative disease, such as in HER2-positive tumors?

In addition to checkpoint inhibition, there are a number of immunotherapeutic strategies being tested in solid tumors. These agents include—but are not limited to—drugs modulating OX40, LAG-3, and GITR. Finally, there are a number of checkpoint inhibitor trials ongoing in operable breast cancer, either along with neoadjuvant cytotoxic therapy or in patients with residual disease who have not responded to neoadjuvant chemotherapy. While we are awaiting these studies, the IMpassion130 trial has changed the treatment landscape in metastatic triple-negative PD-L1–positive breast cancer. ■

Dr. Kalinsky is Assistant Professor of Medicine, Columbia University Irving Medical Center, NewYork–Presbyterian Hospital, New York. Dr. Hershman is Professor of Medicine and Epidemiology, Columbia University Irving Medical Center, NewYork–Presbyterian Hospital, New York.

DISCLOSURE: Dr. Kalinsky reported (immediate family member) stock and other ownership interest in Array BioPharma and Novartis; and he is a consultant/advisor with AstraZeneca, Ipsen, Pfizer, Amgen, Novartis, Eisai, bioTheranostics, and Lilly and is on the speakers bureau for Lilly. Dr. Hershman reported no conflicts of interest.

REFERENCES

1. Schmid P, Adams S, Rugo HS, et al: Atezolizumab and nab-paclitaxel in advanced triple-negative breast cancer. N Engl J Med 379:2108-2121, 2018.

2. Emens LA, Lois S, Rugo HS, et al: IMpassion130: Efficacy in immune biomarker subgroups from the global, randomized, double-blind, placebo-controlled, phase III study of atezolizumab + nab-paclitaxel in patients with treatment-naive, locally advanced or metastatic triple-negative breast cancer. 2018 San Antonio Breast Cancer Symposium. Abstract GS1-04.