Barclays: Merck’s personalized cancer vaccine phase III data: Clear recurrence benefit, but commercial potential still depends on durability

What changed with the phase III data?

Among 1137 patients with resected stage IIB to IV melanoma, the two-year recurrence-free survival rate was 82.2% in the INT plus Keytruda group, versus 70.0% in the placebo plus Keytruda group; the hazard ratio for recurrence or death was 0.58, representing a 42% reduction in relative risk.
Two-year distant metastasis-free survival rates were 88.5% and 80.2%, respectively, with a hazard ratio of 0.62. However, the hazard ratio for this measure in the phase II primary analysis was 0.347; phase II had a smaller sample and wider confidence intervals and did not include stage II patients. Phase III overall survival data remain immature.
On safety, the proportions experiencing grade three or higher adverse events were 23.6% in the combination group and 20.3% in the control group; the proportions discontinuing both study drugs because of adverse events were 6.5% and 5.3%, respectively.

The recurrence-free survival curves largely overlapped during the first six months, with a hazard ratio of 0.99; after six months, it fell to 0.42. INT requires the selection of neoantigens based on a patient’s tumor and the activation of T cells. In the trial, the median time from randomization to the first INT injection was 64 days.
The phase II recurrence-free survival hazard ratio improved from 0.56 at the primary analysis to 0.51 at five-year follow-up. Barclays believes the phase III results by time period provide a positive signal on durability, but the hazard ratio after six months cannot be taken directly as overall efficacy; longer follow-up is still needed.

Barclays believes preoperative neoadjuvant treatment and postoperative adjuvant INT are not a zero-sum proposition. One investigator estimated that about 20% of the high-risk melanoma patients they treat are suitable for preoperative treatment; the remaining patients include those with stage II disease and those found after surgery to have sentinel lymph node involvement. This is one physician’s experience, not a formal market-sizing estimate.
INTerpath-001 excluded patients who had received systemic therapy, so its results apply most directly to patients who undergo surgery first. The effect of using INT after preoperative immunotherapy and its efficacy compared with preoperative regimens remain to be studied.

INT is customized to each patient’s tumor characteristics. Across 157 global trial sites, a product was successfully manufactured for more than 99% of patients randomized to INT; however, 17.3% of screened patients could not enroll because of insufficient tumor samples, and the median wait for the first dose was 64 days. Tissue collection and manufacturing turnaround are therefore practical constraints on wider use.
Fever and injection-site pain were more common in the combination group. Experts at the meeting also cited tissue requirements, manufacturing logistics, drug access and cost as considerations for postoperative treatment.

Barclays has incorporated INT into its model, forecasting risk-adjusted global peak sales of 11.4 billion dollars, of which 5.7 billion dollars would accrue to Merck, contributing about 7 dollars per share to the bank’s net present value estimate. The valuation includes contributions from indications beyond melanoma.
In a phase I pancreatic cancer study, all 19 patients evaluable for immune response developed neoantigen-specific T-cell responses; this demonstrates immune activity, but clinical efficacy remains to be studied. Kidney and bladder cancer studies have completed enrollment, and data release depends on the occurrence of prespecified events.
Barclays maintains its Overweight rating on Merck and its 180-dollar price target; its model is also affected by the pace of Keytruda revenue decline and progress in launching new products.