WHAT THE STUDY ACTUALLY SAYS

A PI3K inhibitor shrank tumours in a third of women with a rare ovarian cancer

In a single-arm phase 2 trial, risovalisib produced a confirmed response in 35% of 84 women with PIK3CA-mutated ovarian clear cell carcinoma. There was no comparison group.

An experimental drug that blocks the enzyme PI3K-alpha shrank tumours in about a third of women with a rare, treatment-resistant form of ovarian cancer, in a single-arm phase 2 trial published in The Lancet Oncology [s1]. The response rate cleared the bar the trial had set in advance. But every patient received the same drug, with no comparison group, so the result is an encouraging signal in a cancer with few options — not proof the drug works better than the alternatives.

The drug and the disease

Ovarian clear cell carcinoma is a rare but aggressive subtype of ovarian cancer that behaves differently from the more common forms and is hard to treat once platinum-based chemotherapy stops working [s1]. Mutations in the gene PIK3CA — which switches on the PI3K-alpha signalling pathway that drives cell growth — are common in this cancer, which makes the pathway an obvious drug target [s1]. Risovalisib, carried in the trial registry under the code CYH33, is an oral selective PI3K-alpha inhibitor [s1][s2].

What the trial did

This was a single-arm, phase 2 trial conducted at 27 centres in China and 12 in Japan [s1]. Eligible participants were adults aged 18 or older, in reasonable general health (an Eastern Cooperative Oncology Group performance status of 0 or 1), with recurrent or persistent PIK3CA-mutated ovarian clear cell carcinoma who had received one to three previous lines of systemic treatment [s1]. Risovalisib was given orally at 40 mg daily in 21-day cycles until the disease progressed or side effects became unacceptable [s1]. The primary endpoint was the objective response rate — the share of patients whose tumours shrank by a defined amount — judged by a blinded independent review committee using standard RECIST criteria, among treated patients with a confirmed PIK3CA hotspot mutation and measurable disease [s1]. The trial was designed to count as positive only if the lower edge of the 95% confidence interval for that response rate exceeded 8% [s1]. It is registered as NCT05043922 [s2].

Between 6 October 2021 and 31 January 2024, 93 patients were enrolled and treated [s1]. Their median age was 49 years (interquartile range 43–58), all were female, 52 (56%) were enrolled in Japan and 41 (44%) in China, and 63 (68%) had already had two or more lines of therapy [s1]. The median follow-up was 14.5 months [s1].

What it found

The confirmed objective response rate was 35% — 29 of 84 evaluable patients — with a 95% confidence interval of 25% to 46% [s1]. Because the lower bound of 25% sits well above the prespecified 8% threshold, the trial met its primary endpoint [s1].

The side-effect profile reflected the drug's mechanism. The most common severe (grade 3 or worse) treatment-related events were rash (30 patients, 32%), high blood sugar (28 patients, 30%) and weight loss (seven patients, 8%) [s1]. Treatment-related serious adverse events occurred in 36 patients (39%), most often rash (eight, 9%), lung inflammation known as pneumonitis (six, 6%) and reduced appetite (four, 4%) [s1]. There were no treatment-related deaths [s1].

How to read it

The design sets the ceiling on what can be concluded. This was a single-arm study: there was no control group, so the 35% response rate cannot be compared against what the same patients would have achieved on another treatment or none [s1]. Response rate also measures tumour shrinkage, not how long patients live or stay free of progression — the outcomes that ultimately decide a cancer drug's value, and which this trial was not built to settle.

The safety signals are worth naming plainly. High blood sugar and rash are predictable, on-target consequences of blocking PI3K-alpha, and here they were severe in roughly a third of patients; serious treatment-related events affected 39% [s1]. That is a meaningful toxicity burden for a drug whose survival benefit is not yet established. The trial was also funded by the drug's developer, Haihe Biopharma, and is still ongoing [s1].

What to watch

The questions that matter next are whether the responses translate into longer progression-free and overall survival, and whether a randomised trial can confirm a benefit over existing options in this rare cancer. The investigators themselves frame risovalisib as a "potential" option and suggest combination strategies to build on the activity seen here [s1]. For now the honest summary is that a targeted drug cleared a prespecified bar in a cancer that badly needs options — a genuine signal, held back from being a verdict by the single-arm design.

This article describes clinical-trial results and is not medical advice. Decisions about cancer treatment are for patients and their treating oncologists.

Sources

  1. Risovalisib for PIK3CA-mutated recurrent or persistent ovarian clear cell carcinoma: a single-arm, multicentre, phase 2 trial — The Lancet Oncology , November 10, 2026
  2. A Study to Evaluate the Efficacy and Safety of CYH33 in Patients With Recurrent/Persistent Ovary Clear Cell Carcinoma (NCT05043922) — ClinicalTrials.gov

More on

Related coverage