The denominator is doing all the work here. Two partial responses among 28 patients is a 7.1% response rate [1], and nobody opens seven new sites [3] on 7.1%. But Aprea has publicly detailed two responding patients, both with PPP2R1A-mutated uterine or endometrial cancer, one dosed at 150 mg and one at 220 mg [15][13], and uterine or endometrial patients were 8 of the 28 enrolled [18]. If both responses came from inside that group of eight, the rate within it is 25% [2] and the trial-wide figure is really describing a mixture of tumor types.
How many of the 28 carried a PPP2R1A mutation is not stated in the poster [17]. Without that count, 25% is a ceiling drawn around a subgroup whose true size is unknown, which is precisely the number the expansion is designed to produce. Aprea's stated aim is to characterize activity in biomarker-defined populations with a mechanistic rationale for WEE1 inhibition [6], and CEO Oren Gilad told GEN the goal is to identify who responds most strongly so that population can be carried into the next phase [11].
The accrual arithmetic is the reason Q4 is the date to hold them to. Six to 10 patients a month [3] means the study adds as many patients in roughly three to five months as it had enrolled in total by the May 6 cutoff [4]. The planned cohort of at least 50 patients with uterine serous carcinoma or Cyclin E-overexpressing platinum-resistant ovarian cancer [7] is about 1.8 times the entire reported dose-escalation population [5]. So the Q4 update should be the first version of this dataset where a mutation-selected group has a denominator worth quoting.
Two cautions on what that update can and cannot settle. It will still be single-arm. In patients with a median of three prior lines of treatment [21], a 50% lesion reduction is hard to attribute to anything but the drug, which is why the two partial responses carry more weight than the six stable diseases; stable disease in this setting has no comparator, and the source reports no durations [19]. And the CA-125 declines, 87% in one case and more than 90% in the other [13][15], are pharmacodynamic readouts. Gilad's own framing is that the next phase is about overall survival [12].
The combination arms are where the safety story gets its real test. Aprea plans to use dose levels that already showed single-agent activity and cleared safety review [9], layered onto standard of care in HPV-positive head and neck cancer and colorectal cancer [8]. The poster's authors concluded manageable safety with no substantial myelosuppression [20], which is easier to sustain as a single agent than on top of chemotherapy.
My read: this is a defensible use of dose-escalation money, conditional on one thing. Whether the tripling bought clarity or just speed depends on whether the Q4 presentation reports response and duration by mutation status, rather than another pooled advanced-solid-tumor rate.
