Sofe-M posts a high response rate in platinum-sensitive ovarian cancer
Eli Lilly has put an early clinical marker in the ground for platinum-sensitive ovarian cancer. In a phase 1 cohort combining sofetabart mipitecan, or Sofe-M, with bevacizumab, the company reported a 79.3% response rate among 33 patients whose disease had returned at least six months after their last platinum dose. Complete responses were seen in 20.7% of the cohort, a signal that stands out in a setting where durable tumor shrinkage can translate into meaningful time off chemotherapy and, in some cases, a better chance of carrying forward into later lines of treatment with less cumulative toxicity.
The response pattern matters because this is a population already selected for platinum sensitivity, a biologic state that often implies retained DNA-damage vulnerability. Sofe-M is an FRα-directed antibody-drug conjugate, and the regimen pairs that targeting strategy with bevacizumab, a VEGF inhibitor that can alter tumor vasculature and potentially improve drug delivery inside the tumor microenvironment. The combination is designed to do more than simply add two active agents together; it aims to intensify tumor killing through a biologically coherent pairing.
Activity held up across FRα expression and prior PARP exposure
The response rate looked consistent across FRα expression levels and prior exposure to PARP inhibition. Among the three patients with FRα expression between 0% and 24%, the response rate was 66.7%, but the sample is too small to treat that figure as a stable subgroup estimate. The broader point is more important: activity did not appear to collapse in patients with lower FRα staining, and the response rate was nearly identical in patients who had and had not previously received a PARP inhibitor.
That detail carries practical weight because PARP inhibitor exposure is common in contemporary ovarian cancer care. Two-thirds of participants had previously received a PARP inhibitor, including agents such as Lynparza, and about 10% had received AbbVie’s FRα-directed ADC Elahere. Prior treatment pressure can narrow options by selecting for resistant clones, changing antigen expression, and reshaping DNA repair dependencies. Sofe-M still produced a robust response signal in that context, suggesting the drug can retain activity even after several layers of prior therapy.
The durability data are still immature, but early treatment continuation looks encouraging. Almost 80% of responders remained on treatment at the data cutoff, and almost 70% of patients had been treated for at least nine months. Median follow-up for progression-free survival was 10.2 months, and the study has not yet reached median PFS. In a phase 1 setting, those numbers do not establish comparative efficacy, yet they do suggest the initial responses are not evaporating immediately after scan confirmation.
A durable response in this setting is not a trivial event.
Safety profile shows hematologic toxicity without an obvious stop signal
The safety profile was dominated by familiar chemotherapy-adjacent toxicities. Grade 3 or worse treatment-emergent adverse events most commonly included neutropenia at 42.4% and anemia at 24.2%. Across all grades, nosebleed and nausea were the most frequent adverse events, each affecting 72.7% of patients. Two patients discontinued treatment because of adverse events.
Those findings fit with the payload biology. Sofe-M uses a TOPO-1 inhibitor, and that class often brings marrow suppression along with gastrointestinal toxicity. The bleeding signal is more difficult to parse from the available data, but bevacizumab can contribute to mucosal bleeding and vascular fragility, making nosebleed a plausible on-target companion effect in a regimen that already carries hematologic burden. Lilly’s investigators described the treatment as well-tolerated, and the discontinuation count suggests the regimen remained deliverable for most participants in this small cohort.
For a first-look dataset, the key question is not whether every toxicity disappeared; it is whether efficacy emerged without an unmanageable safety ceiling. On that measure, the profile appears workable, though the phase 3 program will have to show that the combination can stay effective and tolerable in a broader, more heterogeneous population.
Phase 3 competition is tightening around FRα-targeted disease
Lilly is not entering a vacant field. AbbVie already has Elahere, approved in platinum-resistant ovarian cancer in 2022, and is running a phase 3 study in platinum-sensitive disease with a target population enriched for high FRα expression. Genmab launched a phase 3 PSOC study of rinatabart sesutecan in April, and AstraZeneca recently advanced torvutatug samrotecan into phase 3 testing in the same disease setting. All of these trials are enrolling patients who relapse after platinum-based chemotherapy, but Lilly and some of its competitors are taking a broader approach than AbbVie’s high-FRα strategy.
Payload choice may shape how this market sorts itself. Elahere carries a tubulin inhibitor, while the ADCs from AstraZeneca, Genmab, and Lilly use topoisomerase I inhibitors. In ovarian cancer, where patients may move from one targeted therapy to another, the payload can become part of the resistance story. Sequential use of TOPO-1 ADCs may be less attractive if resistance emerges within that class, which raises the stakes for whichever agent reaches the market first and establishes itself as the default option in platinum-sensitive relapse.
The competitive pressure is especially intense because the clinical bar in PSOC is different from the bar in platinum-resistant disease. Patients with platinum-sensitive relapse often live long enough to receive multiple subsequent therapies, so trial teams need not only a response signal but also enough durability and tolerability to keep patients on treatment while preserving later options. That combination of efficacy and sequencing value is where early market leadership can translate into a durable commercial advantage.
What the early dataset can and cannot support
The phase 1 readout gives Lilly a credible argument that Sofe-M deserves a place in the PSOC race. A 79.3% response rate, 20.7% complete response rate, activity across prior PARP inhibitor exposure, and a tolerable safety profile are enough to justify continued development and to keep attention on the ongoing phase 3 FRAmework-01 program. The trial is enrolling patients whose disease responded to platinum and stayed controlled for at least six months, while control cohorts receive platinum-based doublet chemotherapy plus bevacizumab.
What remains unproven is whether the phase 1 signal will persist when the treatment is tested against established platinum-based therapy in a larger randomized study. The follow-up remains short for a disease where durability often determines whether an early response becomes a meaningful clinical advance. FRα expression, prior PARP exposure, and TOPO-1 payload class all add layers of complexity that can look straightforward in a small cohort and much less forgiving once the population broadens.
For now, Lilly has shown enough to stay in the conversation, and perhaps enough to make rivals take the program seriously. The next readout will have to show whether Sofe-M can hold its own when the comparison is no longer an early cohort, but the current standard of care in a randomized phase 3 trial.