Kura Oncology Inc. CEO Troy Wilson told BioWorld that, based on powerful interim data with tipifarnib (often called "tipi" by researchers) in relapsed or refractory peripheral T-cell lymphoma (PTCL), to amend the ongoing phase II study – which has been rejiggered once already – into a registrational experiment is "a possibility. I think that would not be our preference, but you could do that."
Physicians who treat T-cell lymphoma represent "a pretty small community of docs," Wilson said. "They all know each other. Once we get buy-in from them and buy-in from [the FDA], it's probably cleaner and faster to run a new trial. But all options are on the table," he said, thanks to results that put Kura "standing on the side of the angels." In the fourth-line setting, "you would expect zero responses and patients would maybe stay on study for a few weeks. We are driving patients to complete responses and on to first or second transplants," he said. "That's remarkable."
With the farnesyl transferase inhibitor tipi, San Diego-based Kura turned up continuing antitumor activity and a manageable safety profile in advanced patients with angioimmunoblastic T-cell lymphoma (AITL) as well as non-AITL PTCL. The multicenter, single-arm, open-label study was designed to determine efficacy, safety and biomarkers of activity. At first, patients were enrolled without selection. Then, based upon molecular characterization of the initial subjects, the study was amended to include two expansion cohorts: patients with AITL, an aggressive form of T-cell lymphoma often characterized by high levels of CXCL12 expression and patients with PTCL who lack a single nucleotide variation (rs2839685 A>G) in the 3'-untranslated region of the CXCL12 gene.
As of the May 24 data cutoff, 50 patients with a median number of three prior regimens have been enrolled in all stages of the study. The primary efficacy endpoint was achieved in each of the AITL and CXCL12+ expansion cohorts. Sixteen patients were treated in the AITL cohort and 15 in the CXCL12 SNP+ cohort. Among the 11 evaluable patients in the AITL extension cohort, three achieved a complete response (CR) and two achieved a partial response (PR), for an objective response rate (ORR) of 45% (31% ORR on a modified intent-to-treat basis, mITT). Among the 12 evaluable patients in the CXCL12+ expansion cohort, three achieved a CR and two achieved a PR, for an ORR of 42% (33% ORR by mITT). Two of the five responders in the CXCL12+ expansion cohort were AITL patients.
When all AITL patients (N=23) and all PTCL not otherwise specified (PTCL-NOS) with available rs2839695 data and absence of the 3'UTR variant (n=17) enrolled in all portions were taken into account, ORRs were 53%/39% (PPS/mITT) for AITL and 20%/18% for CXCL12 SNP+ PTCL-NOS. Thirty-four patients had available gene expression data. Patients with a high ratio of CXCL12 expression to its receptor CXCR4 (N=17) experienced an ORR of 47% and a clinical benefit rate of 82% (CR+PR+SD) with tipi.
Next-generation sequencing of 16 AITL patients revealed a high rate of mutation/variation (50%) of the killer cell immunoglobulin-like receptors, including killer immunoglobulin receptor (KIR) 3DL2, the company said. KIR3DL2 mutation at C336R was concurrent with Q386E and was associated with outcome from tipi therapy. Four of the eight KIR3DL2 C336R/Q386E patients achieved a CR, two achieved a PR and two achieved stable disease for a CR rate of 50%, an ORR of 75% and a clinical benefit rate of 100%. Furthermore, high KIR3DL2 mutant variant allele frequency KIR3DL2 was predictive of complete response to tipi in AITL. Tumors with KIR3DL2 mutations expressed low levels of CXCL5 and its receptor CXCR1 and CXCR2, a potential mechanism of resistance to tipi.
Getting the GIST of precision medicine
Tipi was generally well-tolerated, with adverse events consistent with its known safety profile. The most frequently observed treatment-related adverse events (grade ≥ 3) were hematology-related, including thrombocytopenia, neutropenia, leukopenia, anemia, febrile neutropenia and lymphopenia. Results were shared at the European Hematology Association (EHA) annual congress in Amsterdam.
In 2015, Kura completed a reverse merger and closed a $60 million private placement with a syndicate of top venture funds almost at the same time it executed an agreement with Johnson & Johnson (J&J) unit Janssen Pharmaceutica NV, of Beerse, Belgium, for an exclusive license to develop and commercialize tipi in multiple oncology indications. (See BioWorld, March 30, 2015.)
"When we in-licensed this drug, J&J had treated 5,000 patients," Wilson said. "They had seen anecdotal evidence of activity. They had seen tumor shrinkage in a bunch of different tumors, but they didn't have any kind of unified hypothesis about why that was," because J&J lacked the necessary sequencing capacity to explore further. "We figured out that you could treat HRAS mutant tumors and we have a registrational study ongoing in head and neck cancer, but that didn't help us understand why Janssen and its collaborators saw activity in leukemia, lymphoma, breast cancer and pancreatic cancer."
When the "secret weapon" patient database is viewed through the CXCL12 lens, Wilson said, "lo and behold. If the CXCL12 gene is knocked out, everywhere you looked you started to see clinical activity. If you'd known that ahead of time, you would have designed the trial very differently." (See BioWorld, Feb. 8, 2018.)
Wainwright analyst Joseph Pantginis in a report called the new data "crucial in progressing Kura's clinical goal of deploying tipi in targeted cancer populations showing defined biomarkers, hence, supporting a personalized targeted cancer approach in oncology with improved chances of success." SVB Leerink's Jonathan Chang sounded excited, too. He estimated relapsed/refractory PTCL could be worth, at peak, about $140 million in revenue in the U.S. "That being said, the EHA data could be viewed as providing validation to the CXCL12 biomarker thesis that has potentially broader implications in larger indications such as pancreatic cancer, diffuse large B-cell lymphoma (DLBCL), acute myeloid leukemia (AML), and others."
Regarding the relatively small PTCL opportunity, Wilson pointed to gastrointestinal stromal tumors (GIST) as a comparator. "Ten or 15 years ago, no one had ever heard of GIST," he said. "Now you've got GIST patients coming out of the woodwork. This is always the case. When you have an effective drug, the patients find you."
Further efforts in other tumor types will be carried out. Adding "AML, DLBCL, and pancreatic cancer together, and let's just say we take 20-30% [of the market], that's 20,000 to 25,000 patients every year in the U.S. that we could benefit," he said.
Shares of the company (NASDAQ:KURA) closed Friday at $18.62, up 10 cents.