Neoadjuvant talimogene laherparepvec (T-VEC) led to a significant improvement in the 1-year recurrence-free survival rate in patients with resectable advanced melanoma compared with surgery alone, according to a randomized trial presented at the 2019 ASCO Annual Meeting.
Reinhard Dummer, MD
Neoadjuvant talimogene laherparepvec (T-VEC; Imlygic) led to a significant improvement in the 1-year recurrence-free survival (RFS) rate in patients with resectable advanced melanoma compared with surgery alone, according to a randomized trial presented at the 2019 ASCO Annual Meeting.1
At 1 year, 33.5% of patients who received preoperative T-VEC plus surgery remained recurrence free, as compared with 21.9% of patients who had surgery only (HR, 0.73; 80% CI, 0.56-0.93;P= .048). A sensitivity analysis that excluded positive surgical margins as an RFS event produced a larger difference in favor of neoadjuvant therapy (HR, 0.63; 80% CI, 0.47-0.83;P= .024).
More patients randomized to the oncolytic immunotherapeutic were alive at 1 year (95.9% vs 85.8%; HR, 0.47; 80% CI, 0.27-0.82;P= .076), but the difference did not reach statistical significance, as reported at the 2019 ASCO Annual Meeting.
“Neoadjuvant T-VEC resulted in a higher pathologic complete response (pCR) rate in resectable melanoma than that observed by the overall clinical response rate and may account for the higher R0 resection rate in the T-VEC arm than the surgery-alone arm,” Reinhard Dummer, MD, of the University Hospital of Zurich in Switzerland, and colleagues concluded in a poster presentation. “Clinical response is not a predictor of pCR to neoadjuvant T-VEC treatment. The primary analysis of 2-year recurrence-free survival is expected later in 2019.”
The findings constituted a follow-up to previously reported data, which showed a pCR rate of 21% with T-VEC plus surgery and an objective response rate of 14.7%.2The results presented at ASCO 2019 involved 150 patients with high-risk resectable stage IIIB-IVM1a melanoma. The patients were randomized to immediate surgery or intralesional T-VEC, followed by surgery at week 13.
The primary endpoint was RFS. Secondary endpoints included RFS and OS at 2, 3, and 5 years; overall tumor response; pCR in the T-VEC arm; rates of R0 surgical resection; local RFS; regional RFS; and distant metastasis-free survival.
Investigators at 9 sites in nine countries enrolled patients. The intention-to-treat (ITT) analysis included all 150 patients. An efficacy analysis included 57 patients who received at least one dose of T-VEC and had surgery and 69 patients who had immediate surgery. A safety analysis included 73 patients who received at least one dose of T-VEC and 69 patients who had immediate surgery.
The patient groups had no substantive differences in baseline characteristics. In the ITT population, the median patient age was 63.5 years (range, 34-85) in the T-VEC arm (n = 76) and 59.0 years (range, 21-85) in the control arm (n = 74). Overall, 64% of patients were male, 87% of patients had an ECOG performance status of 0, and 13% had an ECOG performance status of 1.
In the efficacy analysis, 13 of 57 patients in the T-VEC arm had a pCR. By ITT analysis, the pCR rate was 17.1% (13 of 76) in the T-VEC arm. More patients in the T-VEC arm had R0 resection status (56.1% vs 40.6%;P= .082). Rates of R1 resection were 42.1% and 55.1%, and R2 resection rates were 1.8% and 4.3% in the T-VEC and surgery-alone arms, respectively.
Investigator-assessed clinical response in the T-VEC arm was 13.2% (10 of 76), including 5 of 13 patients who had pCR and 5 of the 63 who had less than pCR. The disease control rate (response plus stable disease) was 40.8% by ITT analysis, including 11 of 13 patients who had pCR and 20 of 63 patients who had less than pCR after T-VEC treatment. Dummer noted that clinical response to T-VEC at week 12 did not correlate with pCR status.
By ITT analysis, neoadjuvant T-VEC reduced the 1-year recurrence hazard by 27% (80% CI, 0.56-0.93). Median RFS had yet to be reached in either treatment arm. Before the start of follow-up, RFS events had occurred in 56.6% of the T-VEC arm (no surgery in 23.7% and lack of R0 resection in 32.3%) and 60.8% of the immediate-surgery arm (5.4%, 55.4%, respectively).
After excluding less than R0 surgical resection as an RFS event, the 1-year RFS increased to 55.8% in the T-VEC group and 39.3% in the surgery-only arm. The 16-point absolute difference translated into a 37% reduction in the hazard for recurrence in favor of the T-VEC arm (80% CI 0.47-0.83).
The 1-year OS favored the T-VEC arm but the difference did not achieve statistical significance (HR, 0.47; 80% CI 0.27-0.82;P= .076). After a median follow-up of 20.4 months, the median OS had yet to be reached in either treatment group.
Treatment-emergent adverse events in the T-VEC arm were consistent with previously reported data. The most commonly observed treatment-emergent adverse events were flu-like symptoms.
Postoperative adverse event rates were 33.3% in the T-VEC group and 46.4% in the immediate-surgery group. Serious adverse events (SAEs) occurred in 14.0% of the T-VEC group and 2.9% of the immediate-surgery arm. Grade 3 SAEs in the T-VEC group consisted of 2 cases of cellulitis and 1 case each of anembryonic gestation, cholecystitis, device occlusion, influenza, and wound infection. In the immediate-surgery group, grade 3 SAEs consisted of 1 case each of peripheral embolism and wound abscess.
Almost one-fourth of patients had disease progression during T-VEC treatment and did not have surgery, which can be interpreted in two ways, said ASCO invited discussant Joshua M. V. Mammen, MD, PhD, University of Kansas Medical Center.
“Did we lose the opportunity for surgery in patients who perhaps could have benefited from surgical resection?” he asked. “Perhaps, more likely, did we improve patient selection for surgery, really selecting only those patients who might have benefited from surgery?”
“Additionally, it’s interesting that the R0 rate was higher in the T-VEC arm, suggesting a benefit for making the resection easier.”