January 27, 2021 - The combination of the polo-like kinase 1 inhibitor, onvansertib, with FOLFIRI and bevacizumab was found to have preliminary activity and favorable tolerability when used as a second-line treatment for patients with KRAS-mutated metastatic colorectal cancer.
The combination of the polo-like kinase 1 (PLK1) inhibitor, onvansertib, with FOLFIRI and bevacizumab (Avastin) was found to have preliminary activity and favorable tolerability when used as a second-line treatment for patients with KRAS-mutated metastatic colorectal cancer (mCRC), according to results from a phase 1b/2 study (NCT03829410) presented during the virtual 2021 Gastrointestinal Cancers Symposium.1
Forty-two percent of patients (n = 5) evaluated achieved a partial response (PR) with the regimen, including 4 confirmed PRs. Moreover, 1 of these patients went on to receive curative surgery. Sixty-seven percent of patients (n = 8) experienced durable responses that ranged from 6 months to 13 months. Because the combination was found to be well tolerated, the dose level of 15 mg/m2 will be examined further as the recommended phase 2 dose (RP2D).
“Clinical responses were observed across different KRAS variants, including the most common mutations [in CRC]. Patients [who] achieved a PR showed the greatest decrease in plasma mutant KRAS after 1 cycle of therapy,” lead study author Daniel H. Ahn, DO, consultant in the Division of Hematology/Oncology of the Department of Internal Medicine at Mayo Clinic, said during an oral presentation of the data. “Phase 2 of the study will further assess the safety and efficacy of onvansertib in combination with FOLFIRI and bevacizumab, as well as [evaluate] the value of KRAS assessment through liquid biopsy to predict treatment response.”
On average, KRAS is mutated in approximately 50% of patients with CRC, and RAS-targeted therapies have failed, with most of KRAS mutations considered to be “undruggable,” noted Ahn. As such, effective second-line options are needed for these patients, where the median overall survival (OS) is less than 12 months, with an overall response rate of about 5% and a progression-free survival (PFS) of 5.7 months.
PLK1 is a key regulator of mitosis and is overexpressed in CRC; this overexpression is linked with a poor clinical prognosis. PLK1 inhibition has been found to have synthetic lethality with mutant KRAS in CRC cells. Onvansertib, an oral and highly selective PLK1 inhibitor, has demonstrated synergistic and antitumor activity when used in combination with irinotecan and 5-fluorouracil (5-FU) in KRAS-mutated xenograft mouse models; this provided the preclinical rationale for the study presented during the meeting, according to Ahn.
To be eligible for enrollment on the trial, patients had to have metastatic and unresectable CRC, KRAS mutations in the primary tumor or metastasis, have progressed on or were intolerant to oxaliplatin-based chemotherapy, and who experienced disease progression on less than 6 months of first-line therapy. Patients had to be negative for BRAF V600E mutation and microsatellite instability high/deficient mismatch repair status.
In phase 1b portion of the trial, onvansertib was given in combination with FOLFIRI and bevacizumab in successive cohorts of 3 patients, at doses of 12 mg/m2, 15 mg/m2, 18 mg/m2 on days 1 through 5 of two 14-day cycles. Primary end points of the trial included objective response rate (ORR) in patients who received at least 1 cycle of treatment, while secondary end points included PFS and reduction in KRAS allelic burden assessed by liquid biopsy.
Six patients were treated at dose level 0 (onvansertib 12 mg/m2), 3 patients were treated at dose level +1 (onvansertib 15 mg/m2) and 6 patients were treated at dose level +2 (onvansertib 18 mg/m2). Of these patients, 2 remain on treatment at dose level +2 and 1 patient remains on treatment at dose level 0.
The median age of patients included on the study was 60 years, and 53% (n = 8) of them were male. Of the patients who participated in the study, 53% (n = 8) had an ECOG performance status of 1, 20% (n = 3) presented with liver-only metastasis and 40% presented with liver and other metastases. Sixty percent of patients (n = 9) presented with more than 2 metastatic organs, and 67% (n = 10) had received prior bevacizumab.
Additional results from the study showed that 10 patients had a KRAS variant detected by digital droplet PCR at baseline, and clinical responses were observed across different KRAS variants, including the 3 most commonly observed in CRC. The greatest decreases in KRAS mutant allelic frequency (MAF) after 1 cycle of treatment were observed in patients who achieved a PR (range, -78% to -100%). Patients with PR and stable disease tended to have lower on-treatment KRAS MAF than those who had early progressive disease.
The most common treatment-emergent adverse effects (AEs) experienced with the combination included neutropenia (n= 11), fatigue (n = 11) and nausea (n = 9); most of these effects were grade 1 and 2 in severity, although neutropenia had the highest instance of AEs that were grade 3 or higher (n = 4, grade 3; n = 4, grade 4).
Four patients had dose-limiting toxicities (DLTs) attributed to the 5-FU bolus. One patient who received onvansertib at the dose level of 12mg/m2 experienced grade 4 neutropenic fever, while 3 patients who received the agent at the dose level of 18 mg/m2 reported grade 4 neutropenia.
Currently, 3 additional patients are being enrolled to receive onvansertib at a dose level of 15 mg/m2 to further explore the safety of the agent at that dose level. Overall, only 9% of all AEs reported were grade 3 or 4, and no major unexpected toxicities were determined to be associated with onvansertib. Grade 3 or 4 toxicities were managed by dose delay, growth factor support, or discontinuation of the 5-FU bolus. Notably, no patients went off trial because of neutropenia.
The phase 2 portion of the study will further examine the safety and efficacy of onvansertib in combination with FOLFIRI and bevacizumab. Investigators will also evaluate the value of KRAS assessment through liquid biopsy to predict treatment response.