Milestones In Basket Trials: Novartis' Ambitious SIGNATURE Program

Novartis' tissue-agnostic SIGNATURE trial uses eight of the company's drugs in an ambitious set of Phase II protocols that will be an important test of whether basket trial designs are useful in corporate oncology R&D programs.

Novartis AG's SIGNATURE trial is the most extensive company-sponsored basket trial yet attempted, testing eight of the company's drugs in patients selected for a wide array of biomarkers in a histology-independent study.

Roche's VE BASKET trial may be the first significant basket trial from a sponsor company, but it limited its scope to one drug – Zelboraf (vemurafenib) (see related story, (Also see "Tissue-Agnostic Approach To Cancer Drug Development Takes A Hit" - Pink Sheet, 14 Sep, 2015.)). The biggest and broadest basket trials, the American Society for Clinical Oncology's TAPUR study and the National Cancer Institute's NCI-MATCH trial, can use drugs from multiple sponsors because they are being conducted by non-commercial entities.

While Novartis will be looking for signals of efficacy in the SIGNATURE data, the company also is "trying to establish a working model" of a basket trial platform that could be used for "signal-finding very early on in a drug's development," Novartis Senior VP August Salvado, the head of Novartis U.S. Oncology Clinical Development and Medical Affairs, told "The Pink Sheet" in an interview.

The basket trial model is designed to trawl for signs of activity in marginal indications. Conventional randomized clinical trials will be undertaken for populations where actionable mutations are widespread or preclinical data are especially encouraging. "We did not want to get in the way of our development programs, so we basically excluded the most promising areas," Salvado said.

The company has been rolling out new clinical cohorts in SIGNATURE steadily since 2013 (see chart). An update provided at the American Society for Clinical Oncology annual meeting earlier this year covered 368 patients dosed through January 2015. Eight cohorts had been completed for the first three drugs – the PI3K inhibitor buparlisib, the multikinase inhibitor dovitinib and the MEK inhibitor binimetinib.

Novartis will keep growing the SIGNATURE program. "The next level would be to look at combinations rather than single agents," Salvado said. "Many investigators are anxious to move on and combine targeted therapies."

"One or two" combinations are expected to enter SIGNATURE testing in 2016, he noted; the company will look at preclinical data and "the most prevalent" driver mutations to pick a "broadly applicable" regimen.

Novartis' SIGNATURE Program
Modular tissue-agnostic, genetic alteration-specific Phase II protocols to assess the clinical benefit rate (stable disease or better) of targeted therapies in patients with relapsed or refractory pathway-activated tumors.
Actionable Mutations	Treatment	Size	Status
CDK4/6, Cyclin D1/3, p16	Ribociclib (LEE011)	n=90	Recruiting; started 8/2014.
ALK, ROS1	Zykadia (ceritinib)	n=70	Recruiting; started 9/2014.
FGFR	BGJ398	n=90	Recruiting; started 7/2014.
PTCH, SMO	Erivedge (sonidegib)	n=10	Started 2/2014, ended 3/2015.
BRAF V600	Encorafenib (LGX818)	n=12	Active, not recruiting; started 12/2013.
RAS, RAF, MEK	Binimetinib (MEK162)	n=110	Started 10/2013. Data from 90 patients presented at ASCO 2015; preliminary activity seen in AML, ovarian and thyroid cancer.
FGFR, PDGFR, other pathways inhibited by dovitinib	Dovitinib (TKI258)	n=80	Started 8/2013. Data from 73 patients presented at ASCO 2015; preliminary activity seen in ovarian cancer.
PI3K	Buparlisib (BKM120)	n=145	Started 3/2013. Data from 142 patients presented at ASCO 2015; preliminary activity seen in vaginal cancer and head and neck squamous cell carcinoma.
Sources: Company materials, ASCO abstract, clinicaltrials.gov