EFS Findings for Neoadjuvant Pembrolizumab/Chemotherapy Suggest a Potential New Standard of Care in Early TNBC, Leading to FDA Approval

Peter Schmid, MD, PhD, FRCP

Neoadjuvant pembrolizumab (Keytruda), in combination with platinum-containing chemotherapy induced a statistically significant, clinically meaningful increase in pathologic complete response (pCR) and event-free survival (EFS) in patients with high-risk, early-stage triple-negative breast cancer (TNBC), according to findings from the phase 3 KEYNOTE-522 trial (NCT03036488).¹

Additionally, a favorable trend has been noted in overall survival (OS), although follow-up is ongoing; the safety profile was also promising with no new safety concerns noted. “I’m pleased to say that KEYNOTE-522 met both its primary end points,” said Peter Schmid, MD, PhD, FRCP, when presenting results from a prespecified interim analysis of the study during a European Society for Medical Oncology Virtual Plenary session in July. “We’ve now also shown that the addition of pembrolizumab to neoadjuvant chemotherapy followed by adjuvant pembrolizumab showed a statistically significant and, in my opinion, clinically meaningful improvement in event-free survival rates.” Previously, neoadjuvant chemotherapy was the standard-of-care approach for early-stage TNBC, which is typically associated with a poor prognosis and early recurrences, explained Schmid, professor of cancer medicine and group leader of the Centre for Experimental Cancer Medicine at Barts Cancer Institute at Queen Mary University of London in the United Kingdom. Following the results of this trial, the FDA approved the use of pembrolizumab in combination with chemotherapy as neoadjuvant treatment for patients with high-risk, early-stage TNBC, followed by single-agent adjuvant pembrolizumab after surgery (see page 14).²

KEYNOTE-522 was a randomized, double-blind study that enrolled patients with newly diagnosed stage II or III (T1c, N1-2 or T2-4, N0-2) TNBC with an ECOG performance status of 0 or 1. A total of 1174 patients were randomized 2:1 to receive either pembrolizumab and chemotherapy or chemotherapy as neoadjuvant therapy followed by surgery and pembrolizumab or placebo as adjuvant therapy.

The neoadjuvant phase consisted of 2 treatment phases for the chemotherapy in which carboplatin (area under the curve [AUC] 5 every 3 weeks or AUC 1.5 every week) plus paclitaxel (weekly 80 mg/m²) was given for 12 weeks (cycles 1-4), followed by doxorubicin (60 mg/m² every 3 weeks) or epirubicin (90 mg/m² every 3 weeks) plus cyclophosphamide (600 mg/m² every 3 weeks) for 12 weeks (cycles 5-8) regardless of treatment arm.

Pembrolizumab was administered at 200 mg every 3 weeks in both the neoadjuvant and adjuvant phases. Radiation therapy was also given as needed in the adjuvant phase.

The primary end points of the study were pCR (defined as the absence of cancer, ypT0/Tis ypN0) by local pathologist assessment in the intention-to-treat (ITT) population and EFS by investigator assessment in the ITT population. Secondary end points included OS, efficacy outcomes in the PD-L1–positive population, and safety. Exploratory analyses also looked at the distant progression or distant recurrence- free survival and EFS in subgroups.

Between the 2 arms, patients’ ages ranged from 22 to 80 years, and the majority had an ECOG performance status of 0 (87%), PD-L1– positive disease (83%), T1/T2 tumor size (74%), and lymph node involvement (52%).

Interim Analyses Results

An earlier interim analysis of the study showed that 64.8% of patients in the pembrolizumab- containing arm achieved a pCR compared with 51.2% of patients in the chemotherapy- alone arm, representing a difference of 13.6 percentage points (P = .00055). The EFS rate at 18 months was 91.3% in the investigational arm vs 85.3% in the control arm (HR, 0.63; 95% CI, 0.43-0.93; P = .0089).³

After a median follow-up of 39.1 months in both arms, the EFS rate at 36 months was 84.5% in the pembrolizumab-plus-chemotherapy arm and 76.8% in the chemotherapy- plus-placebo arm (HR, 0.63; 95% CI, 0.48-0.82; P = .00031). Schmid noted that the EFS analysis had reached statistical significance as of this later analysis.

“The hazard ratio has been consistent across all interim analyses observed so far, and considering the pattern of triple-negative breast cancer and the disease recurrence pattern, it is, in my opinion, highly unlikely to change over time,” Schmid said.

The most common adverse event (AE) in both arms was distant recurrence at 7.7% and 13.1% in the investigational and control arms, respectively, followed by local recurrence in 3.6% and 4.4% of patients. Fifteen patients (1.9%) died in the pembrolizumab arm vs 6 (1.5%) in the placebo arm. Schmid commented that of the patients who experienced local recurrence, many of them (13 in the pembrolizumab arm and 9 in the placebo arm) experienced subsequent distant recurrence as well.

All subgroup analyses favored the pembrolizumab arm in terms of EFS, with significant benefit seen in the PD-L1–negative (HR, 0.48) and T1/T2 (HR, 0.51) subgroups for the addition of pembrolizumab.

There was a greater difference between the treatment arms in EFS in patients who did not achieve a pCR. “The type or quality of pathologic response seems to be altered by the addition of pembrolizumab to chemotherapy. Patients with residual disease at the time of surgery still have substantially higher rate of event-free survival if they were treated with pembrolizumab,” Schmid said.

In patients who did achieve a pCR, the EFS rates at 36 months were 94.4% with pembrolizumab and 92.5% with placebo; in patients who did not achieve a pCR, the EFS rates were 67.4% and 56.8% in the pembrolizumab and placebo arms, respectively.

The rate of distant progression or distant recurrence-free survival at 36 months was 87.0% in the pembrolizumab-and-chemotherapy arm vs 80.7% in the chemotherapy-and-placebo arm (HR, 0.61; 95% CI, 0.46-0.82).

OS rate at 36 months was 89.7% with pembrolizumab and 86.9% without pembrolizumab (HR, 0.72; 95% CI, 0.51-1.02; P = .03214). The prespecified boundary of 0.00086 was not reached as of this analysis.

Safety Profile

The rates of treatment-related adverse events (TRAEs) were generally similar between the 2 treatment arms in the neoadjuvant and adjuvant phases combined. Overall, TRAEs were observed in 98.9% of patients in the pembrolizumab arm and 99.7% in the chemotherapy- alone arm. Grade 3 to 5 TRAEs were observed in 77.1% of patients in the pembrolizumab arm and in 73.3% of patients in the chemotherapy arm. Events leading to death were reported in 0.5% of patients in the pembrolizumab arm and in 0.3% of patients in the chemotherapy arm. TRAEs led to drug discontinuation in 27.7% and 14.1% of patients in the pembrolizumab arm and the chemotherapy-alone arm, respectively.

The most common TRAEs were nausea (63.2% with pembrolizumab vs 63.0% with chemotherapy), alopecia (60.2% vs 56.6%, respectively), anemia (54.8% vs 55.3%), neutropenia (46.9% vs 47.6%), and fatigue (42.1% vs 38.8%).

In the adjuvant phase alone, TRAEs of any grade were observed in 53.7% of patients treated with pembrolizumab compared with 48.6% of patients treated with placebo. Grade 3 to 5 TRAEs were reported in 6.3% and 2.7% of patients in the pembrolizumab and placebo arms, respectively; death due to TRAEswas reported in 0.3% and 0% of patients, respectively. Approximately 4.3% of patients discontinued pembrolizumab due to TRAEs and 1.8% discontinued placebo.

The most common TRAEs in the adjuvant phase were arthralgia (8.5% with pembrolizumab vs 6.9% with placebo), rash (6.0% vs 2.4%, respectively), pruritus (5.1% vs 2.1%), asthenia (4.6% vs 5.1%), diarrhea (4.1% vs 3.3%), and fatigue (3.4% vs 4.8%).

Immune-mediated AEs and infusion reactions in both phases combined were more commonly observed in the pembrolizumab- and-chemotherapy arm than in the chemotherapy- alone arm (any grade, 43.6% vs 21.9%, respectively; grade 3-5, 14.9% vs 2.1%). The most common AEs were infusion reactions (18.0% with pembrolizumab vs 11.6% without), hypothyroidism (15.1% vs 5.7%, respectively), severe skin reactions (5.7% vs 1.0%), and hyperthyroidism (5.2% vs 1.8%).

In the adjuvant phase, immune-mediated AEs and infusion reactions were reported in 10.2% of patients in the pembrolizumab arm vs 6.0% of the chemotherapy-alone arm; events were grade 3 to 5 in 2.9% and 0.3% of patients, respectively.

Context in Terms of Other (Neo) Adjuvant TNBC Trials

Giuseppe Curigliano, MD, PhD, provided further context for the results of the KEYNOTE- 522 trial during the plenary session. Curigliano, the associate professor of medical oncology at the University of Milan and head of the Division of Early Drug Development at the European Institute of Oncology in Italy, said that the data from the KEYNOTE-522 trial were practice changing but that many questions now emerge with these data.

The EFS results were highly consistent with the results from the phase 3 GeparNuevo trial (NCT02685059), in which the 3-year EFS rate was 85.6% with the addition of immunotherapy with durvalumab (Imfinzi) to anthracycline/ taxane-based neoadjuvant chemotherapy in patients with early TNBC vs 77.2% in patients treated with chemotherapy and placebo (HR, 0.48; 95% CI, 0.24-0.97; P = .0398).⁴

Compared with the phase 3 IMpassion031 trial (NCT03197935), both studies showed that the addition of immunotherapy to neoadjuvant chemotherapy improves pCR rates in the ITT patient populations. In IMpassion031, the pCR rate was 57.6% in patients treated with atezolizumab (Tecentriq) plus nab-paclitaxel (Abraxane) and anthracycline–based chemotherapy in patients with early-stage TNBC vs 41.1% in patients who received chemotherapy and placebo, for a difference of 16.5 percentage points (P = .0044).⁵

There was no difference in pCR rates according to disease burden. In both trials, the rates of pCR were similar in patients with node-negative and node-positive disease treated with the addition of immunotherapy, but there was difference in the chemotherapy-alone arms.

Curigliano contemplated whether neoadjuvant pembrolizumab should be given to all patients with high-risk, early-stage TNBC regardless of PD-L1 expression because a benefit was seen in both PD-L1–positive and PD-L1– negative patients in the KEYNOTE-522 trial.

When analyzing the results according to PD-L1 expression levels (by combined positive score [CPS]), a more significant benefit was observed for the addition of pembrolizumab in patients with CPS equal to or greater than 10 and CPS equal to or greater than 20 tumors. Among patients with CPS equal to or greater than 10 (50% of patients), the pCR rate was 77.9% in those treated with pembrolizumab vs 59.8% in those not treated with pembrolizumab and 81.7% vs 62.5%, respectively, in patients with CPS equal to or greater than 20 (32% of patients). This aligned with the difference in pCR seen according to PD-L1 expression (positive vs negative) in the IMpassion031 trial. Curigliano noted that these patients may need to be better identified.

He also suggested that the greatest benefit for added pembrolizumab was seen in patients with a worse prognosis, such as those who did not achieve a pCR, and he stressed that these patients also need to be better identified to find who may benefit from added immunotherapy the most in the neoadjuvant setting.

_REFERENCES:

_{1. Schmid P, Cortes J, Dent R, et al. KEYNOTE-522: phase III study of neoadjuvant pembrolizumab + chemotherapy vs. placebo + chemotherapy, followed by adjuvant pembrolizumab vs. placebo for early- stage TNBC. Ann Oncol. 2021;32(9):1198-1200. doi:10.1016/j.annonc.2021.06.014}

_{2. FDA approves Keytruda (pembrolizumab) for treatment of patients with high-risk early-stage triple-negative breast cancer in combination with chemotherapy as neoadjuvant treatment, then continued as single agent as adjuvant treatment after surgery. News release. Merck. July 27, 2021. Accessed August 19, 2021. https://bit.ly/3BNscgP}

_{3. Schmid P, Cortes J, Pusztai L, et al; KEYNOTE-522 Investigators. Pembrolizumab for early triple-negative breast cancer. N Engl J Med. 2020;382(9):810-821. doi:10.1056/NEJMoa1910549}

_{4. Loibl S, Schneeweiss A, Huober JB, et al. Durvalumab improves long-term outcome in TNBC: results from the phase II randomized GeparNUEVO study investigating neoadjuvant durvalumab in addition to an anthracycline/taxane based neoadjuvant chemotherapy in early triple-negative breast cancer (TNBC). J Clin Oncol. 2021;39(suppl 15):506. doi:10.1200/JCO.2021.39.15_suppl.506}

_{5. Mittendorf EA, Zhang H, Barrios CH, et al. Neoadjuvant atezolizumab in combination with sequential nab-paclitaxel and anthracycline-based chemotherapy versus placebo and chemotherapy in patients with early- stage triple-negative breast cancer (IMpassion031): a randomised, double-blind, phase 3 trial. Lancet. 2020;396(10257):1090-1100. doi:10.1016/S0140-6736(20)31953-X}