Pfizer's $1B protein degradation bet yields 3% response rate in Arvinas' midphase breast cancer trial

Pfizer’s $1 billion bet on Arvinas’ protein degradation technology has delivered phase 2 data. After a publication snafu, Arvinas posted the breast cancer results ahead of schedule—and revealed one partial response among 35 patients at its phase 3 dose. 

Arvinas planned to present the data Dec. 8. However, the San Antonio Breast Cancer Symposium “incorrectly published the abstract, omitting a key safety data table, and inadvertently released the corresponding full data presentation” on its website on Monday, according to the biotech. In response to the incident, Arvinas released its full-data presentation on Tuesday, two weeks ahead of schedule.

The data raise questions about the efficacy of ARV-471, a leading light in the hot protein degradation field that Pfizer secured co-development rights to last year in a deal that cost it $1 billion upfront. Shares in Arvinas fell 16% on the day the data leaked. 

In the phase 2 expansion cohort, investigators treated 71 patients with ER-positive, HER2-negative breast cancer with one of two doses of ARV-471. All patients had previously received a CDK4/6 inhibitor such as Pfizer’s Ibrance. Among the 44 subjects with measurable disease, two patients, one in each dose cohort, had confirmed partial responses. The response at the higher dose happened in a patient with an ESR1 mutation. 

Another five participants, all of whom had ESR1 mutations, had unconfirmed partial responses but were listed as having stable disease in the presentation. One responder was excluded from the 44 patients in the efficacy analysis because of a “lack of complete set of target lesion measurements on-study.” Arvinas plans to test the lower, 200-mg dose in its phase 3 clinical trial. At that dose, Arvinas saw one confirmed partial response and three unconfirmed partial responses, meaning the response rate is 11%, at best. 

The biotech used clinical benefit rate (CBR) as its primary endpoint, enabling it to include patients with 24 weeks of stable disease in its analysis. Overall, the CBR came in at 38%, with the low- and high-dose arms performing comparably. The CBR was higher in patients with mutant ESR1, a marker Arvinas will stratify by in phase 3, clocking in at 51.2% in the overall subpopulation and 47.4% at the 200-mg dose. 

In the overall population, median progression-free survival (PFS) was 3.5 months in the 200-mg cohort and 3.7 months across both dose levels. Again, narrowing the focus to people with mutant ESR1 boosted the figures, with PFS at 200 mg in the genetically defined subpopulation rising to 5.5 months. 

One positive for Arvinas is that it achieved the results in a heavily pretreated patient population. The phase 3 trial will treat a slightly different population. All participants will have received one prior line of a CDK4/6 inhibitor but, unlike in the phase 2, none will have tried AstraZeneca’s selective estrogen receptor degrader Faslodex before.

Faslodex, which will be given to the control arm in the phase 3, was used by 56% of participants in the midphase study prior to treatment with ARV-471. Similarly, unlike in the phase 2, patients in the phase 3 cannot have taken chemotherapy for advanced disease.