Addition of Neoadjuvant Pembrolizumab Improves EFS in Resectable Melanoma

Sapna Patel, BA, MD

Compared with adjuvant pembrolizumab (Keytruda) alone, the addition of neoadjuvant pembrolizumab significantly improved event-free survival (EFS) outcomes for patients with stage III-IV melanoma with a hazard ratio of 0.58 (95% CI, 0.39-0.87; P = .004), according to data from the phase 2 SWOG S1801 trial (NCT03698019).¹

The landmark 2-year EFS rate was 72% in the neoadjuvant arm (n = 159) compared with 49% in the adjuvant arm (n = 154). Tumor-related events—including disease progression, adjuvant therapy delays, and melanoma recurrence—occurred in 38 patients in the neoadjuvant arm and 67 patients in the adjuvant arm.

Twelve patients had disease progression that precluded surgery and 10 of patients had residual disease or developed a metastasis prior to adjuvant therapy in the neoadjuvant arm. Comparatively, 17 patients in the adjuvant arm experienced residual disease or developed metastasis prior to starting therapy.

“The administration of anti–PD-1 blockade as therapy before surgery in the neoadjuvant setting induces an immune response from a larger population of T cells that reside in the bulk of the tumor,” explained Sapna Patel, BA, MD, in a presentation of the data.

She explained that by inhibiting the PD-1/PD-L1 immune checkpoint prior to surgery that induces a systemic anti-tumor response, which occurs before resection of the tumor bed. Researchers believe that this approach leaves behind larger numbers of anti-tumor T cells that can be activated and systemically circulated to recognize and attack any micrometastatic disease.

The phase 2 trial randomly assigned 345 patients 1:1 to either the adjuvant arm where they were given 18 cycles of 200-mg intravenous (IV) pembrolizumab every 3 weeks, or to the neoadjuvant arm, where they were given the same amount prior to surgery but also given 15 cycles of 200-mg pembrolizumab every 3 weeks after surgery.

Between the 2 arms, patient characteristics were similar with the median age of patients above 60 in both arms of the study. Moreover, over 80% of patients in both arms had low or normal lactate dehydrogenase levels, but the arms were more even when looking at the status of the patient’s disease. The slight majority in both arms were patients with stage IIIC disease, no ulceration, and BRAF negative.

Pembrolizumab in the neoadjuvant arm also benefited EFS in all key subgroups in comparison with the adjuvant group. In particular, patients aged 66 years old or older had an HR of 0.4 (95% CI, 0.21-0.74), male patients had an HR of 0.48 (95% CI, 0.29-0.8), patients with stage IIIIC disease had an HR of 0.4 (95% CI, 0.22-0.75), and patients with BRAF-mutated disease had an HR of 0.44 (95% CI, 0.21-0.91).

Overall survival was also looked at in the study, but data was not mature at the time of presentation with only 36 deaths reported on the trial (HR, 0.63; 95% CI, 0.32-1.24; P = .18).

Nine participants in the neoadjuvant arm achieved a complete radiographic response and 59 had a partial response to therapy. Forty-two patients had an increase in their target lesions but 30 of those patients were still able to receive surgery. One of the patients with a complete response declined surgery and had not had a recurrence of disease after 31.5 months of follow-up. Upon local review, 21% of patients in the neoadjuvant arm also achieved a pathologic complete response.

“Events occurring during the surgical period occurred in 21 participants in the adjuvant arm, most occurring at a post-operative tumor assessment radiology scan,” added Patel, director of the Uveal Melanoma Program and Melanoma Fellowship Program at The University of Texas MD Anderson Cancer Center in Houston, Texas. “In the neoadjuvant arm, there were 14 events during the neoadjuvant period occurring at a post-pembrolizumab pre-surgery scan, and 14 events in the surgical period at a post-operative tumor assessment radiology scan.”

According to Patel, there were no new safety signals for pembrolizumab observed in this study and rates of grade 3 or 4 treatment-related adverse events (TRAEs) were similar between the study arms. The most common TRAE in both arms was infection in the skin or wound with 2 patients in the adjuvant arm and 3 in the neoadjuvant arm, but in the adjuvant arm, 4 patients experienced maculo-popular rash compared with 1 patient in the neoadjuvant arm. Researchers also saw that pembrolizumab given in the neoadjuvant setting did not increase AEs during the peri-operative period.

As of the time of the presentation, 41 patients in the adjuvant arm were still on therapy along with 43 patients in the neoadjuvant arm. Fewer patients discontinued treatment in the neoadjuvant arm than the adjuvant arm at 59 vs 71 patients, respectively.

“I think it's hard to argue with the event-free survival that you're seeing, but we do see that there are some patients who get neoadjuvant therapy and adjuvant therapy who recur quite early,” Patel said in a discussion of disease resistance at the event. “On both [arms of the study], there was a percentage of patients who actually recurred prior to the initiation of adjuvant therapy. So, we do need to consider upfront therapy for resectable patients and then also consider for patients in whom the therapy is not working what are our alternative strategies or markers for that resistance.”

Reference

Patel S, Othus M, Prieto V, et al. Neoadjvuant versus adjuvant pembrolizumab for resected stage III-IV melanoma (SWOG S1801). Ann Oncol. 2022;33(suppl 7):S808-S869. doi:10.1016/annonc/annonc1089