Clinical study design
This single-center, nonrandomized, dose escalation, open-label, investigator-initiated phase I study enrolled patients at Mayo Clinic between patients between October 2017 and September 2019. Eligible patients were those with advanced or metastatic solid malignancy for which no standard treatment option existed that would confer clinical benefit and who had life expectancy of 12 weeks or greater at the time of evaluation;≥18 years of age with biopsy-confirmed malignancy; ECOG PS of 0 or 1; adequate bone marrow and organ function; QTcF of <450msec; and measurable disease per RECIST. All their side effects from previous treatments or surgery must have resolved to grade 1 or better according to CTCAEv4.03. A negative pregnancy test within 7 days prior to study drug was required in women of childbearing potential; contraception practices were required for women (or men sexually active with women) who were of child-bearing potential. All patients had to be able to swallow the ribociclib capsule.
The study excluded patients who had received anti-cancer chemotherapy, immunotherapy, radiotherapy, or investigational agents within 28 days prior to registration or limited field radiation for palliation or major surgery ≤14 days prior to registration or if ≥25% bone marrow was irradiated. Also excluded were patients who had impaired GI function or disease that may affect absorption of study drugs; baseline neuropathy of > grade 2; active, clinically serious infection or known history of HIV infection; other serious uncontrolled medical conditions; or hypersensitivity to any of the excipients of ribociclib. If they had known central nervous system (CNS) involvement they were excluded unless they were at least 4 weeks from prior therapy completion (clinically stable CNS disease without steroids and/or enzyme-inducing anti-epileptic medications for brain metastasis were allowed).
Prohibited concomitant medications included strong inducers or inhibitors or CYP3A4/5, drugs with narrow therapeutic window and predominantly metabolized through CYP3A4, medications known to cause QT prolongation or induce Torsades de Pointes and herbal preparations.
Two study phases were planned: a dose escalation phase and an expansion phase. The dose escalation phase was based upon the standard modified Fibonacci 3+3 study design and projected to accrue up to 30 patients. The four escalation dosing levels (Supplemental Table 1) were prespecified based on the anticipation that the main DLT would be myelosuppression due to it being a common AE of both study agents. The starting dose level used reduced standard clinical doses of both study agents. Each subsequent level escalated the dose of one of the study agents. The rationale used to determine the anticipated acceptable maximum combined dosing for the final escalation dose level (DL4) was based on a standard clinical dosing of gemcitabine (1000 mg/m2) and 800 mg/day of ribociclib (one dose level greater than standard clinical dosing of 600 mg/day). The dose level at which MTD was determined would be identified as the RP2D and used for the expansion phase. The expansion phase planned to enroll an additional 20 patients to reasonably evaluate the PK and other correlative science endpoints as well as DLTs to elaborate the decision regarding the appropriateness of the choice for the MTD from the dose escalation phase. Investigators and patients were not blinded to study treatment.
All participants granted written informed consent according to federal and institutional guidelines. The trial was conducted with approval of Mayo Institutional Review Board. The study is registered with ClinicalTrials.gov with the identifier NCT03237390.
Study treatment
Treatment cycles were 21 days in length. Gemcitabine was administered intravenously on cycle days 1 and 8 and ribociclib taken orally once daily on days 8-15, followed by a 7-day rest. In order to minimize the inter-patient variability of PK assessments, participants took their first cycle dose of ribociclib immediately before gemcitabine infusion on day 8. Novartis provided ribociclib. The initial cohort in the dose escalation phase started at Dose Level 1 and subsequent patient cohorts were enrolled based on the safety and tolerability of the preceding dose level cohort as outlined in Supplemental Table 1.
Adverse events (AE) were graded according to the National Cancer Institute Common Toxicity Criteria version 4 (NCI CTCAE v4.0). When patients experienced Grade 3 or Grade 4 treatment related toxicity or intolerable Grade 2 toxicity despite optimal supportive care, treatment might be delayed and/or dose reduced. In the event of multiple toxicities, dose modification was based on the worst toxicity observed. For any AE ≥ Grade 2, ribociclib was dose interrupted until recovery to Grade ≤ 1, except for Grade 2 anemia or neutropenia for which no dose adjustment was required. When a patient required dose modification of ribociclib due to protocol-guided AE, the dose was reduced by 200 mg per day. Ribociclib dose reductions to the next lower dose level were required if a patient experienced a recurrence of Grade 3 thrombocytopenia, a Grade 3 neutropenia that took > 7 days to resolve, first Grade 4 neutropenia, first Grade 3 febrile neutropenia, or any Grade 3 nonhematological AE. Ribociclib was discontinued if a patient developed a Grade 4 anemia, febrile neutropenia, or non-hematological toxicity. Gemcitabine was postponed on Day 1 without dose adjustment on resumption for thrombocytopenia with platelet count < 100 x 109/L until platelets recovered to above that threshold; neutropenia Grade 2/3 until recovery to ≤ Grade 1; anemia Grade 3 until recovery to Grade ≤ 2; and non-hematological AE Grade 2 until recovery to Grade ≤ 1. Gemcitabine was dose reduced after recovery from Grade 4 thrombocytopenia and neutropenia or Grade 3 febrile neutropenia. Gemcitabine was discontinued if a patient developed Grade 4 febrile neutropenia or anemia.
Endpoints
The primary objectives were to describe the dose-limiting toxicities (DLT) and identify the MTD and recommended Phase II dose (RP2D) of the combination of ribociclib and gemcitabine in patients with advanced solid tumors. The MTD was defined as the maximum dose level at which ≤ 1/6 patients had DLT in the dose escalation phase. All patients who received the combination of study drugs within the first cycle were considered evaluable for toxicity.
DLT for this trial were defined as grade 4 neutropenia > 7 consecutive days; grade 4 thrombocytopenia or grade 3 with bleeding; grade 3 or 4 febrile neutropenia; QTc interval ≥ 501ms on ≥ 2 separate EKGs; cardiotoxicity or troponin ≥grade 3 or clinical signs of cardiac disease such as unstable angina or myocardial infarction; vomiting ≥ grade 3 over 48 hours despite optimal anti-emetic therapy; diarrhea ≥ grade 3 over 48 hours despite optimal anti-diarrheal therapy; bilirubin ≥ grade 2 for over 7 consecutive days or grade 3; ALT ≥grade 2 with a ≥grade 2 bilirubin elevation of any duration in absence of liver metastases, ALT≥ grade 3 for more than 4 consecutive days; grade 4 ALT or AST; grade 4 serum alkaline phosphatase >7 consecutive days; serum creatinine ≥grade 3; any non-hematologic events ≥grade 3 (excluding alopecia; grade 3 fatigue <5 days, grade 3 fever or infection without neutropenia < 5 days duration; grade 3 laboratory abnormalities responsive to oral supplementation or deemed by the investigator to be clinically insignificant). Persistent, intolerable treatment related toxicities which delayed treatment for >14 days, and failure to receive at least 80% of the scheduled doses due to treatment related toxicity (except when treatment delay is due to sub-optimally managed nausea, vomiting or diarrhea) were also considered DLT. If a patient did not have a DLT but could not complete at least 80% of ribociclib and 2 doses of gemcitabine, then they would be replaced. Safety evaluations were completed throughout the study and a follow-up safety evaluation included AE assessment and review of concomitant medications and occurred 30 days (+/- 3 days) after the last dose of study drug or until resolution of any drug related toxicities.
The secondary objectives were to describe pharmacokinetics, antitumor activity of the combination of ribociclib and gemcitabine, and correlative biomarker analysis. Secondary endpoints of efficacy included response rate (RR), progression free survival (PFS), and short-term survival. Tumors were measured at baseline and prior to or on Cycle 3 Day 1, then prior to or on Day 1 of every other cycle thereafter. RECIST 1.1 was used to assess and document RR in terms of complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). PFS was defined as time from enrollment to disease progression determined either clinically or radiographically. Three months after patients terminated treatment, follow-up short-term survival information was collected without further follow-up afterwards.
Pharmacokinetics
Serial whole blood samples for pharmacokinetic analysis of ribociclib were collected in two 2 mL EDTA tubes during the first cycle of treatment, 30 minutes before and 0.5, 1, 2, 4, 6, and 8 hours after administration of ribociclib on day 8 and day 14 of the first cycle of treatment. Plasma was isolated by centrifugation at 1,500 x g for 10 minutes within 30 minutes following blood collection and aliquots (2 ml each) were placed in ice immediately and then stored frozen at ‑70ºC or below until analyzed. The plasma concentration of ribociclib was determined using a liquid chromatography/tandem mass spectrometry (LC-MS/MS) assay. Quality assurance was maintained by injecting quality control samples each time patient samples were assayed.
Plasma concentration-time data were analyzed by standard non-compartmental methods using the program WinNonlin (Pharsight, Mountain View, CA). The areas under the plasma concentration time curve (AUC) on day 8 and day 14 were calculated by trapezoidal approximation. The accumulation ratio (R) was calculated as the ratio of day 14 AUC0-8h versus the day 8 AUC0-8h. The apparent elimination half-life (t1/2) was calculated as –(0.693*τ)/ln((R-1)/R) where R is accumulation ratio and τ is the dosing interval (24 hours). Since a 24-hour blood sample was not drawn on day 14, the 24-hour plasma concentration after ribociclib administration was estimated to be equivalent to the pre-dose concentration based on the assumption that steady-state was reached on day 14, and the AUC over the 24-hour dosing interval on day 14 (AUCt) was calculated by trapezoidal approximation. Oral steady-state clearance (CLSS/F) was calculated using the equation, CLSS/F = Dose/AUCt, where dose is the administered dose of ribociclib. Standard descriptive statistics were used to summarize plasma ribociclib PK parameters.
Biomarker correlates
Optional tissue samples (archival) when available or biopsy if feasible were collected to evaluate correlation between CDK2/4/6, Cyclin D1 and Cyclin D3amplification, RB and P16 expression in archived and/or biopsied tumor tissue and treatment response. Tissue samples were labeled as “primary”, “metastatic” or “recurrent” upon submission. The tissue could be archived as either a paraffin-embedded tumor block or unstained slides (a minimum of 10 slides, 5 µ thickness with tumor content of 20% or higher as assessed by hematoxylin and eosin).
Statistical analysis
No power analysis was performed as the study was not powered for any endpoints. These data were analyzed descriptively (simple counts, percentages, and median) to assess the primary and secondary endpoints.