For phosphoinositide 3-kinase (PI3K) drug development, "the black eye came very much from the Gilead experience with Zydelig last year," acknowledged Verastem Inc. CEO Robert Forrester, but that bruise was healed by the "totality, consistency and magnitude" of phase III data offered by his firm with duvelisib in chronic lymphocytic leukemia (CLL) at the American Society of Hematology (ASH) meeting in Atlanta.

"I think from the key opinion leaders' perspective, the scientists', the physicians' perspective, PI3K has always been a mechanism of interest," Forrester said. "People recognize that biologically it plays a really important role, and if one could target it effectively, could have material benefit to patients. That's what we're now seeing."

Foster City, Calif.-based Gilead Sciences Inc.'s twice-daily, oral PI3K delta inhibitor, Zydelig (idelalisib), approved with a black-box warning for relapsed follicular B-cell non-Hodgkin's lymphoma and relapsed CLL, ran into trouble in March 2016 when the company halted six trials after deaths in trials prompted an FDA alert. Gilead had been testing the drug as a first-line therapy. (See BioWorld Today, Aug. 29, 2016.)

"They were not prophylaxing their patients, largely," which led some to cases of colitis and opportunistic infections, Forrester said. Such patients – males 72 years old, on average – are especially susceptible because their immune systems are compromised. Verastem, of Boston, has been proactive in that regard, he said, and if duvelisib wins approval, "we want to put in place programs and support for patients, nurses and doctors to make sure they know what to expect and how to handle the side effects," he said. "They're very predictable and manageable. There's nothing surprising in the data we've just presented."

Nothing negatively surprising, that is. Roth Capital analyst Jotin Marango said his firm "remain[s] bullish on Verastem, not only because of the positive readout itself, but importantly because of what these data mean for duvelisib in the commercial landscape of CLL generally and the league of PI3K agents specifically." Duvelisib is a first-in-class oral dual inhibitor of PI3K-delta and PI3K-gamma in the works for CLL/small lymphocytic leukemia (SLL) as well as follicular lymphoma (FL). Verastem plans to submit an NDA in the first quarter of this year in CLL/SLL, and will ask for accelerated approval in FL.

"What must duvelisib do to establish a footprint in relapsed CLL? The answer, in our view, is nothing," Marango wrote in a report, saying the compound "is poised to ride competitive trends manifesting before our eyes, including the aggressive move of ibrutinib [Imbruvica, Abbvie Inc./Janssen Biotech Inc.] to front-line (potentially up to 60 percent of patients), opening second and third lines to oral agents, specifically venetoclax [Venclexta, Abbvie Inc./Roche Holding AG] (which we expect to take the majority share), and PI3K inhibitors. Importantly, as we saw at ASH, even the VIPs have real issues: 24 percent of front-line ibrutinib patients discontinue; 16 percent of relapsed patients on venetoclax cannot reach recommended dose, 83 percent require hospitalization for titration monitoring, and 35 percent discontinue. Meanwhile, Zydelig has flat-lined at $38 million average quarterly sales after its black box. As we see it, even on equal footing with Zydelig, there is sufficient available market share for duvelisib as another PI3K option in relapsed CLL, although after the tolerability profile from the DUO study they are certainly not on equal footing."

Imbruvica is a Bruton tyrosine kinase inhibitor. Venclexta blocks B-cell lymphoma-2 protein.

Forrester called Venclexta "a very good drug, but quite challenging to initiate," with a 12-page dosing guide. "You have to start with a ridiculously low dose," and up-titrate over a five-week schedule. "Many of the patients have to be hospitalized because of the concern over tumor lysis syndrome," he said.

Following 'poster child' ibrutinib's path

In the 319-patient Duo trial, duvelisib monotherapy's safety profile landed nicely in line with previous studies. For duvelisib-treated patients, the median time on treatment was 50.3 weeks (range 0.9-160) vs. 23.1 weeks (range 0.1-26.1) for comparator Arzerra (ofatumumab, Novartis AG). The most common grade ≥3 treatment-emergent hematologic adverse events (occurring in >10 percent of patients) were neutropenia (30 percent) and anemia (13 percent). The most common grade ≥3 nonhematologic treatment-emergent adverse events (occurring in >10 percent of patients) were diarrhea (15 percent), pneumonia (14 percent) and colitis (12 percent). The rate of severe opportunistic infections was 6 percent, including two patients (1 percent) with Pneumocystis jirovecii pneumonia (PJP), neither of whom was on prophylaxis for PJP at the time of the event.

Thirty-five percent of patients discontinued duvelisib treatment due to an adverse event; about 40 percent of patients treated with duvelisib remained on treatment for more than 18 months, with a median total follow-up of nearly two years. Adverse events (AEs) of interest infrequently led to discontinuation of duvelisib treatment (e.g., diarrhea [5.1 percent], colitis [5.1 percent], pneumonitis [1.9 percent], neutropenia [1.3 percent], pneumonia [1.3 percent], transaminase elevations [0.6 percent], and rash [0.6 percent]). Duvelisib treatment-related AEs leading to death (n=4) include general physical health deterioration (n=1); pneumonia staphylococcal (n=2); and sepsis (n=1).

Efficacy proved solid, too, with full results highlighting duvelisib's superiority in all patient subgroups. The experiment met its primary endpoint, with monotherapy hitting a statistically significant improvement in median PFS (mPFS) compared to ofatumumab per a blinded independent review committee (IRC) using iwCLL or revised IWG Response Criteria (modified iwCLL/IWG; 13.3 months vs. 9.9 months, respectively; HR=0.52; p<0.0001), representing a 48 percent reduction in the risk of progression or death.

Similar efficacy of duvelisib was observed regardless of whether patients had the harder-to-treat 17p deletion (del[17p]). The primary outcome of mPFS via IRC review in the del[17p] subpopulation significantly favored duvelisib over ofatumumab (12.7 months vs. 9 months, respectively; HR=0.41; p=0.0011), a 59 percent reduction in the risk of progression or death. By investigator assessment, duvelisib demonstrated an mPFS of 17.6 months, compared to 9.7 months for ofatumumab (HR=0.40, p<0.0001). Duvelisib maintained a PFS advantage in all patient subgroups analyzed as a subset of pre-specified sensitivity analyses.

Good news came, too, on the secondary efficacy outcome of overall response rate (ORR) via IRC assessment according to modified iwCLL/IWG. Results significantly favored duvelisib over ofatumumab (73.8 percent vs. 45.3 percent, respectively; p<0.0001), and reduced lymph node burden >50 percent in most patients vs. ofatumumab (85 percent vs. 16 percent). In the del[17p] subpopulation of patients, ORR was also significantly higher for duvelisib compared to ofatumumab, 70 percent vs. 43 percent, respectively (p=0.0182). Overall survival (OS) in the intent-to-treat population was similar for those randomized to duvelisib and to ofatumumab during the study (HR=0.99; p=0.4807), demonstrating no detrimental effect on OS and was likely due to other available therapies following progression.

Patients who progressed in the DUO study were given the option to enroll in a crossover study to receive the opposite treatment. In the optional crossover study, 89 patients who were previously treated with ofatumumab in DUO and experienced disease progression were subsequently treated with duvelisib monotherapy. As in the parent DUO study, duvelisib showed robust clinical activity, with an ORR of 73 percent, a median duration of response of 12.7 months and an mPFS of 15 months by investigator assessments.

Verastem's goal with duvelisib is to make CLL/SLL into a disease with which patients can "lead a perfectly normal life, taking a pill twice a day," Forrester said, adding that, in late-stage FL, PI3K is "the mechanism of choice for physicians. Ibrutinib and venetoclax generally do not have much activity in those patient populations."

The company hopes to move into early stage FL eventually, he said. In first- or second-line FL, patients may be given Revlimid (lenalidomide, Celgene Corp.) or Rituxan (rituximab, Roche Holding AG/Biogen Inc.). Chemotherapy may be used, but "by the time you get to third line, the patients really can't tolerate it and are not enthusiastic about doing it," he said, calling the FL market size "comparable" to that of CLL.

The CLL/SLL opportunity could grow, Forrester said, likening it to renal cell carcinoma, "a relatively small disease. The market size increased because more patients wanted to try these [newer] therapies. If they are oral, easy to administer, with no hospitalization, no infusions, and have a manageable side-effect profile," then patient adoption is not so difficult to bring about, he said. "Ibrutinib has been the poster child for that, and we believe duvelisib can continue down that path."