Immatics N.V. (Nasdaq: IMTX) reported second quarter results and a business update on August 18, 2026, and used the release to disclose a change in how its lead Phase 3 trial will be analysed. The company, which has operations in Houston, Texas and Tuebingen, Germany, develops therapies directed at the PRAME antigen across cell therapies and bispecific antibodies. The disclosure was furnished to the U.S. Securities and Exchange Commission as exhibits to a filing on EDGAR.
Cash and cash equivalents together with other financial assets stood at $448.2 million, or EUR 393.4 million, as of June 30, 2026, against $534.7 million, or EUR 469.3 million, at December 31, 2025. The company converts at the European Central Bank rate in effect on June 30, 2026, of 1 EUR = 1.1394 USD, and projects cash reach into 2028. The decline reflects research and development spending, partially offset by net proceeds of $24.2 million from an at-the-market offering and by working capital and exchange rate movements.
The SUPRAME redesign
SUPRAME is a global, randomised, controlled, multi-centre Phase 3 trial of anzu-cel, previously called IMA203, as monotherapy against investigator’s choice in patients with unresectable or metastatic melanoma who have already received a PD-1 immune checkpoint inhibitor. The primary endpoint is progression-free survival assessed by blinded independent central review under RECIST v1.1. Key secondary endpoints include overall survival, objective response rate, safety and patient-reported quality of life measures. Anzu-cel holds FDA Orphan Drug Designation and RMAT designation.
Immatics said aggregate progression-free survival events, meaning progressive disease or death, are occurring more slowly than originally modelled. In response, and following interaction with the FDA, the company intends to replace the previously planned interim and final PFS analyses with a single streamlined final analysis based on a lower prespecified number of events, while maintaining 90 percent power for the primary endpoint. Separately it intends to enrol approximately 90 additional patients to raise statistical power on overall survival, bringing the total trial size to approximately 450 patients. The company states that the larger overall survival event requirement does not affect the timing of the final PFS analysis.
Enrollment remains on track to complete the required randomisations by year-end. Topline PFS data are expected in the first half of 2027, followed by a Biologics License Application in 2027.
Analysis: what a slower event rate does and does not tell you
The registry entry for SUPRAME, NCT06743126, lists an estimated enrollment of 360 and an actual start date of January 14, 2025, with primary completion estimated for January 2028 and study completion for October 2031, across 71 listed locations. The trial was recruiting as of the last registry update posted on August 18, 2026. Set that 360 against the approximately 450 the company now describes and the size of the amendment becomes concrete: the trial is being enlarged by roughly a quarter, and the reason given is the secondary endpoint, not the primary one.
An event-driven trial reads out when a set number of events has accumulated, not on a calendar. Fewer events than modelled at a given time means patients in the trial as a whole are progressing more slowly than the assumptions used to size it. That is a statement about the pooled population, both arms together, and it is unblinded to nobody outside the independent review process. It does not identify which arm is contributing the shortfall, and the release makes no claim that it does. Any reading of a slow event rate as evidence about anzu-cel specifically goes beyond what this disclosure establishes.
What the company has done in response is a change to the statistical design. Dropping the interim analysis removes an alpha penalty, which is why a lower prespecified event count can still support 90 percent power at the final analysis. Adding roughly 90 patients does nothing for PFS timing, because those patients enrol late and contribute few events by the PFS cutoff, but it does add exposure time toward the overall survival analysis that follows. The trade the company has made is to give up an early look in exchange for a stronger survival dataset at the end, and it says the changes came out of FDA feedback rather than its own initiative alone.
Chief executive and co-founder Harpreet Singh said the approach “provides the most efficient path to generating definitive data for regulatory approval”. The efficiency claim is testable against a single date. If topline PFS data appear in the first half of 2027 as guided, the removal of the interim analysis cost nothing in time. If randomisations do not complete by year-end, that date moves, and the additional 90 patients become part of the reason.
The quarter itself
Total revenue, all from collaboration agreements, was $10.4 million, or EUR 9.1 million, for the three months ended June 30, 2026, against $5.4 million, or EUR 4.7 million, a year earlier, which the company attributes to a higher proportion of collaboration costs incurred within the quarter. Research and development expenses rose to $71.1 million, or EUR 62.4 million, from $51.4 million, driven by clinical trial activity and SUPRAME in particular. General and administrative expenses were $15.8 million against $14.6 million, with the increase put down to commercialisation preparation.
Net loss narrowed to $71.2 million, or EUR 62.5 million, from $80.1 million a year earlier. The company attributes the improvement mainly to unrealised non-cash foreign exchange losses in the prior-year quarter, and secondarily to higher collaboration revenue, offset in part by SUPRAME costs. That composition matters: the loss narrowed despite operating spending rising, so the improvement is not a reduction in cash burn.
Pipeline data and the addressable population
Updated Phase 1b data for anzu-cel in metastatic melanoma presented at the 2026 ASCO Annual Meeting showed a 56 percent confirmed objective response rate, median duration of response of 14.6 months, median PFS of 6.1 months and median overall survival of 16.2 months, with an overall survival rate of 70 percent at 12 months and 46 percent at 24 months. For IMA203CD8, updated Phase 1 data in gynecologic cancers showed a 63 percent objective response rate and 50 percent confirmed response rate with four complete responses, and data in synovial sarcoma showed a 67 percent response rate and 64 percent confirmed. Explorative analyses on predictors of durable response have been accepted for the ESMO Congress 2026.
Immatics puts the addressable population for anzu-cel’s first target indications, second-line or later advanced cutaneous melanoma and metastatic uveal melanoma, at approximately 9,000 PRAME-positive and HLA-A*02:01-positive patients per year in the United States and the EU5, citing Clarivate for 2025.
The regulatory route runs through a Biologics License Application to the FDA’s Center for Biologics Evaluation and Research, the pathway that covers gene therapy products and somatic cells. CBER issues a licence only when it determines that the data adequately demonstrate safety, effectiveness and quality and that the manufacturing facility meets required standards, and the application must carry the manufacturing and controls detail, the nonclinical reports, the results of all human clinical trials and the proposed labelling. Approval of that application, not the PFS readout alone, is what the commercial infrastructure the company is building depends on.
