Phase I Study of GDC-0425, a Checkpoint Kinase 1 Inhibitor, in Combination with Gemcitabine in Patients with Refractory Solid Tumors

Checkpoint kinase 1 (Chk1) is a serine/threonine kinase that functions as a central mediator of the S and G₂ cell-cycle checkpoints. As a consequence of DNA damage or replication stress, the ATR/Chk1 pathway becomes activated, which leads to phosphorylation and inhibition of CDK1/2 and a transient delay in cell-cycle progression so that DNA can be properly repaired (1–4). Inhibition of Chk1 results in checkpoint failure, and thus an otherwise transient genotoxic insult becomes a cytotoxic event as cells enter mitosis with unrepaired DNA. In addition, inhibition of Chk1 or its upstream regulator ATR in the context of replication stress leads to DNA double-strand breaks and replication catastrophe that result in cell death (5, 6). In cancer cell lines, Chk1 inhibition preferentially enhances the activity of antimetabolite-based DNA-damaging agents, such as gemcitabine, compared with other classes of DNA-damaging agents (7). Greater chemopotentiation, via inhibition of Chk1 or Wee1, has also been observed in cells lacking p53 tumor suppressor activity (7, 8). Thus, targeting Chk1 is a strategy for selectively potentiating the efficacy of chemotherapeutic agents, particularly in tumor cells that lack functional p53 (10–12).

GDC-0425 is an orally bioavailable, highly selective small-molecule inhibitor of Chk1. In preclinical studies, GDC-0425, when administered in combination with gemcitabine, abrogates the S and G₂ checkpoints, resulting in premature entry into mitosis, and mitotic catastrophe (refs. 13, 14; Gazzard, manuscript in preparation). GDC-0425 administration effectively reverses gemcitabine-induced cell-cycle arrest as measured by phospho-CDK1/2 (pCDK1/2) and enhances the levels of γH2AX, a marker of double-stranded DNA breaks, above levels observed with gemcitabine alone, both in vitro and in xenograft models in vivo (Gazzard, manuscript in preparation; ref. 15). In addition, checkpoint abrogation may be more rapid in tumor cells defective in p53 tumor suppressor function (16, 17). In vivo studies suggest that chemopotentiation may be more effective with a defined lag between gemcitabine administration and Chk1 inhibition (9, 18).

Together, these data provided the rationale for investigating GDC-0425 in combination with gemcitabine for the treatment of patients with refractory solid tumors. The primary objectives of this study were to evaluate the safety and tolerability of GDC-0425 and to determine the maximum tolerated dose (MTD) in combination with gemcitabine. We also sought to characterize the pharmacokinetic (PK) properties of GDC-0425 after single and repeat doses and identify a recommended phase II dose and schedule for GDC-0425 in combination with gemcitabine. Other objectives included a preliminary assessment of antitumor activity, a correlation of this activity with known TP53 mutation status, and pharmacodynamic (PD) changes in Chk1 pathway components.

This phase I open-label dose-escalation study of GDC-0425 (supplied by Genentech, Inc.) had 2 treatment arms (Supplementary Fig. S1). Initially, patients received oral GDC-0425 alone daily for 3 consecutive days starting on day −7 during cycle 0. Beginning with cycle 1, patients received intravenous gemcitabine on days 1 and 8 and GDC-0425 daily on days 2 to 4 and 9 to 11 of 21-day cycles. In Arm A, patients received gemcitabine 750 mg/m², and in Arm B, patients received gemcitabine 1,000 mg/m². On the basis of validated preclinical models to establish the starting dose of GDC-0425 in line with standard phase I oncology clinical trials, and predicted half-life, the GDC-0425 starting dose was defined as 60 mg once daily. Because of unexpected toxicity likely related to prolonged GDC-0425 exposures, the study was amended to evaluate a single oral dose of GDC-0425 on day −7 for PK evaluation during cycle 0 followed by 21-day cycles of intravenous gemcitabine on days 1 and 8 and GDC-0425 on days 2 and 9 in cycles ≥ 1. Upon determination of the MTD, an expansion cohort was planned to enroll 6 additional patients to confirm safety and tolerability and to assess PD changes within tumors.

Following a 3 + 3 + 3 dose-escalation design, MTD was defined as the highest dose at which fewer than one third of at least 6 dose-limiting toxicity (DLT)-evaluable patients had a dose-limiting toxic effect during either cycle 0 or 1. A DLT was defined as a study drug-related toxicity occurring during the first 28 days, including grade ≥ 4 thrombocytopenia or anemia, grade ≥ 4 febrile neutropenia or neutropenia lasting > 7 days, grade ≥ 3 elevation of alanine aminotransferase (ALT) or aspartate aminotransferase (AST) lasting > 7 days, total bilirubin or nonhematologic or nonhepatic major organ adverse event (AE). In addition, grade ≥ 3 left ventricular ejection fraction reduction or asymptomatic left ventricular ejection fraction (LVEF) reduction to ≤40% were defined as DLTs. The protocol was amended during enrolment to Arm A so that a delay in day 8 gemcitabine dosing, for example, due to low-grade hematologic toxicities commonly observed with gemcitabine administration alone (e.g., grade 1 or 2 neutropenia or thrombocytopenia), was no longer considered a DLT.

Eligible patients age ≥ 18 years had locally advanced or metastatic solid tumors for which standard therapy either does not exist or has proven ineffective or intolerable. Eastern Cooperative Oncology group (ECOG) performance status of 0 to 1 and adequate hematologic and end organ function and evaluable disease or measurable disease per RECIST v1.1 were required. Patients with more than 2 prior chemotherapy regimens for locally advanced or metastatic cancer, more than 6 cycles of an alkylating or platinum agent, or history of symptomatic congestive heart failure, myocardial infarction, or serious cardiac arrhythmia were excluded.

The protocol was approved by Institutional Review Boards prior to patient recruitment and was conducted in accordance with International Conference on Harmonization E6 Guidelines for Good Clinical Practice. Written informed consent was obtained for all patients before performing study-related procedures in accordance with federal and institutional guidelines. The study was registered on ClinicalTrials.gov (NCT01359696).

All patients who received ≥1 dose of study treatment were included in the safety evaluation and underwent physical examination, laboratory assessments, electrocardiography, and radiographic disease assessments at baseline and throughout the study. ECHO or MUGA scans were performed on cycle 1 day 9 following GDC-0425 administration and at study treatment discontinuation. AEs were graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, v4.0.

In the first cohort (single-agent GDC-0425 for 3 consecutive days in cycle 0), the following PK sampling scheme for GDC-0425 and thiocyanate (SCN) was used in cycle 0: predose and at 0.5, 1, 1.5, 2, 4, and 6 hours after dose on day 1; predose and at 2 hours after dose on day 2; predose and at 0.5, 1, 1.5, 2, 4, 6, 24, and 48 hours after dose on day 3. In subsequent cohorts, PK samples for GDC-0425 and SCN in cycle 0 were predose and at 0.5, 1, 1.5, 2, 4, 6, 24, and 48 hours after dose. During cycle 1 (combination therapy with gemcitabine), samples were collected for GDC-0425 and SCN analysis on days 2 and 9 at predose, 2 and 4 hours after dosing on days 2 and 9. Additional predose samples for GDC-0425 and SCN analysis were collected on day 15 in cycle 1 and on day 8 of subsequent cycles.

Serum samples for gemcitabine PK analysis were collected in cycle 1 on day 1 at 15 and 30 minutes and 1, 2, and 3 hours after start of gemcitabine infusion.

PK parameters were derived from noncompartmental analysis (WinNonlin Professional version 5.2) from the plasma concentration–time profile of GDC-0425. A validated liquid chromatographic-tandem mass spectrometry (LC/MS-MS) method with a lower limit of quantitation of 1.00 ng/mL was used to measure the concentration of GDC-0425 in plasma samples (19). A validated LC/MS-MS with a lower limit of quantitation of 50 ng/mL was used to measure the concentration of gemcitabine in plasma samples. A validated UV/VIS spectrophotometry method with a lower limit of quantitation of 25 μmol/L was used to measure the concentration of SCN in plasma samples (20). GDC-0425 and gemcitabine samples were analyzed at Covance Laboratories and SCN samples were analyzed at QPS.

Best overall response, objective response, and time on study were evaluated by arm and dose level. Objective response was defined as a complete or partial response (PR) confirmed ≥4 weeks after initial documentation, as assessed by the investigator using RECIST v1.1 for patients with measurable disease. Patients with no post-baseline tumor response assessment were considered nonresponders. Time on study was defined as time from first GDC-0425 dose to study discontinuation.

TP53 mutations were evaluated in archival tumor tissue by next-generation sequencing (Asuragen or Expression Analysis). The Asuragen assay is based on 32 amplicons covering all TP53 coding regions (21). The Expression Analysis assay is based on targeted hybridization capture covering coding regions of more than 200 genes including TP53 (22). The variant frequency cutoff used for calling TP53 mutations using either assay was 10%.

PD changes in the Chk1 pathway marker pCDK1/2 and Ki-67 were evaluated in formalin-fixed and paraffin-embedded (FFPE) serial tumor biopsies collected before treatment, after GDC-0425 alone (∼24 hours after administration), after gemcitabine alone (∼48 hours after administration), and after the combination of GDC-0425 and gemcitabine (∼48 hours after gemcitabine and 24 hours after GDC-0425) when feasible. FFPE biopsy specimens were sectioned at 4 μm onto Superfrost Plus slides. Parallel sections were stained on Ventana Benchmark XT and Discovery XT machines with antibodies to pCDK1/2 (clone EPR2233Y; Epitomics) and Ki-67 (clone SP6; Lab Vision/NeoMarkers). pCDK1/2 signal intensity in viable tumor cells was evaluated on a 4-point scale (0 to 3+) and then a pCDK1/2 H-score was calculated on the basis of the fraction of positive cells at each intensity (23). Ki-67 positivity in viable tumor cells was scored from 0% to 100%. To account for tumors with different proliferative fractions, pCDK1/2 levels by H-score were normalized by Ki-67 percent positivity.

A PD biomarker study was conducted in a mouse tumor model under conditions that simulate GDC-0425 exposure in the clinical setting. TP53-mutant HT-29 colorectal tumor xenografts were established by subcutaneous injection of 5 × 10⁶ tumor cells (100 μL in 1:1 HEPES-buffered saline:Matrigel; BD Biosciences) into the left flank of female NCR nude mice (Taconic). When tumor volumes reached approximately 300 to 450 mm³, animals were randomized into balanced cohorts and administered saline or 120 mg/kg gemcitabine by intraperitoneal injection. MCT (0.5% w/v methylcellulose/0.2% v/v Tween-80 in reverse osmosis water) or GDC-0425 (50 mg/kg suspension in MCT) was administered by oral gavage 16 hours later. FFPE tumor samples were collected at 40 hours (40 hours after gemcitabine and 24 hours after GDC-0425) for analysis by pCDK1/2 immunohistochemistry as described above. All in vivo studies were approved by Genentech's Institutional Animal Care and Use Committee and adhere to the NIH Guidelines for the Care and Use of Laboratory Animals.

No formal hypotheses were tested in this study, and all analyses were descriptive and exploratory. Design considerations were not made with regard to explicit power and type I error but to obtain preliminary safety, PK, and PD information. For the safety analysis and the activity analysis, all patients who received ≥1 dose of GDC-0425 were included. For PD analyses, biomarker changes after gemcitabine alone or the combination of gemcitabine and GDC-0425 were evaluated by using a one-way ANOVA model for xenograft samples and a linear mixed-effect model for patient samples with patient as a random effect. All statistical analyses were carried out in SAS 9.2 and R 3.1.1. The data cutoff was May 29, 2014.

A total of 40 patients were enrolled and received at least one dose of GDC-0425. In Arm A (n = 18; combination with gemcitabine 750 mg/m²), patients initially received 60 mg GDC-0425 for 3 consecutive days during the single-agent run-in and following gemcitabine administration. GDC-0425 administration was abbreviated to 1 day during the single-agent run-in and following gemcitabine dosing on the basis of PK and tolerability (see below). In Arm B (n = 22; combination with gemcitabine 1,000 mg/m²), patients received 60 mg or 80 mg of GDC-0425 for 1 day during the single-agent run-in and following gemcitabine administration. Of the 39 patients who received gemcitabine, 7 patients did not receive cycle 1 day 8 gemcitabine and, on the basis of the protocol-specified dosing schedule, received their second dose of gemcitabine on day 1 of cycle 2. An additional 6 patients discontinued study treatment and did not receive gemcitabine after cycle 1 day 1.

The baseline characteristics of the patient population divided by treatment arm are shown in Table 1. The median age was 56 years with slightly more females than males. Triple-negative breast cancer (TNBC) made up the highest proportion of malignancies (20% of all patients), followed by non–small cell lung cancer (NSCLC) and cancers of unknown primary origin (CUP). Eighty-five percent of patients had prior systemic treatment with a median of 2 prior therapies and 43% had prior radiation. Eighty-five percent (n = 34) of patients were gemcitabine-naïve.

All 18 and 22 patients in Arms A and B, respectively, were evaluable for safety (Table 2). No DLTs were reported in the cycle 0 GDC-0425 single-agent run-in (3 consecutive daily doses or single day dose). In Arm A, 3 patients treated with 3 consecutive daily doses of 60 mg GDC-0425 in combination with gemcitabine 750 mg/m² had protocol-defined DLTs of grade 3 nausea (n = 1), grade 2 pyrexia with associated grade 3 syncope (n = 1), and a patient with both grade 2 neutropenia and grade 4 thrombocytopenia (n = 1). Thus, 3 consecutive daily doses of 60 mg GDC-0425 in combination with gemcitabine 750 mg/m² were not tolerated.

Following protocol amendments, GDC-0425 administration was changed to 1 day dosing and subsequently 60 mg GDC-0425 with gemcitabine 750 mg/m² was tolerated which allowed evaluation of 60 mg GDC-0425 with gemcitabine 1,000 mg/m². With further dose escalation, 80 mg of GDC-0425 with gemcitabine 1,000 mg/m² was found to exceed the MTD with 3 of 6 patients experiencing DLTs of grade 4 thrombocytopenia; 1 patient with grade 4 thrombocytopenia also experienced grade 3 neutropenia that delayed cycle 2, a second protocol-defined DLT. The MTD and recommended phase II dose (RP2D) was defined as gemcitabine 1,000 mg/m² on days 1 and 8 followed by 60 mg GDC-0425 on days 2 and 9 of a 21-day cycle. Of note, all DLTs were reversible, blood counts recovered with treatment interruption, and no DLTs were attributed to GDC-0425 alone.

In cycle 0, AEs were grade 1 and 2 and the most common were nausea (n = 11, 28%), vomiting (n = 8, 20%), and diarrhea (n = 6, 15%). Limited hematologic toxicity was observed in cycle 0 with only 2 patients experiencing grade 1 and 2 anemia. Of the AEs assessed as related to GDC-0425, gemcitabine, or the combination, across both arms, the most common all grade nonhematologic toxicities were nausea (n = 19, 48%), vomiting (n = 18, 45%), fatigue (n = 17, 42%), and pyrexia (n = 16, 40%). Neutropenia (n = 16, 40%) was the most common grade ≥ 3 AE followed by thrombocytopenia (n = 6, 15%). Treatment-related ALT and AST elevations ≥ grade 3 occurred in 10% and 8% of patients, respectively.

Overall, half the patients (n = 20) experienced at least 1 serious AE (SAE) regardless of attribution. Eight of these patients experienced SAEs assessed by the investigator as related to study treatment (GDC-0425 and/or gemcitabine), including 2 patients each who experienced grade 4 thrombocytopenia and grade 3 neutropenia assessed as related to GDC-0425 and gemcitabine. Other SAEs assessed as related to GDC-0425 and/or gemcitabine occurred in 1 patient each and included pyrexia (grade 2), dyspnea, gastroenteritis, gastric ulcer, leukopenia, rash, and ALT, AST, and γ-glutamyl transferase (GGT) increased (all grade 3). No deaths occurred during this study.

In total, 12 patients (30%) were discontinued from GDC-0425 and gemcitabine treatment due to AEs. Of these, 6 patients were withdrawn for AEs deemed to be treatment-related by the investigator [neutropenia concurrent with thrombocytopenia, pyrexia, and dyspnea in Arm A and elevated liver enzymes (ALT, AST, and GGT), fatigue, and rash in Arm B].

Among the 16 patients treated at the RP2D, 6 patients (38%) had AEs that led to a dose reduction of gemcitabine and 3 patients (19%) had AEs that led to a dose reduction of GDC-0425. Of the 7 patients (44%) who experienced at least one SAE, 2 patients (13%) experienced SAEs assessed by the investigator as related to study treatment (GDC-0425 and/or gemcitabine), including 1 patient with grade 3 rash and 1 patient with grade 3 ALT increased, grade 3 AST increased, and grade 3 GGT increased. Five patients (31%) were discontinued from GDC-0425 and gemcitabine treatment due to AEs including 2 patients (13%) who were withdrawn for AEs deemed related to study treatment by the investigator.

PK plasma samples from 39 patients were analyzed for GDC-0425 and gemcitabine. Following oral administration, GDC-0425 was rapidly absorbed with median T_max ranging from 2–4 hours (Table 3, Fig. 1). After reaching peak plasma concentrations, concentrations decreased with a terminal phase half-life of approximately 15 hours (range, 7.52–27.3 hours). Mean C_max following 3 once-daily doses of 60 mg GDC-0425 was approximately 1.6-fold higher than mean C_max following a 60-mg single dose. There was considerable overlap in the range of plasma exposures with 60 mg and 80 mg GDC-0425. Plasma exposures of 60 mg GDC-0425 given in combination with 750 mg/m² and 1,000 mg/m² gemcitabine were consistent. Overall, the exposure and half-life observed in patients were approximately 6- and 2-fold higher, respectively, than what was predicted from nonclinical models. This higher-than-expected exposure was a factor that was considered in changing the dosing schedule from 3 days to 1 day after the first cohort. In addition, both dosing regimens exceeded the predicted threshold for instigating checkpoint failure, a binary event, for approximately 24 hours (with 1 day of dosing) or longer (with 3 days of dosing; Fig. 1).

Gemcitabine plasma concentrations peaked at the end of infusion and declined rapidly thereafter, with a half-life of approximately 0.25 hours (range, 0.172–0.358 hours). Maximum plasma concentrations and area under the curve (AUC) of gemcitabine following 1,000 mg/m² 30-minute intravenous infusion were higher than 750 mg/m² 30-minute intravenous infusion (C_max: 18,100 ± 8,440 ng/mL vs. 11,500 ± 5,810 ng/mL and AUC_inf: 9,110 ± 3,900 h ng/mL vs. 4,860 ± 1,080 hr ng/mL). Overall, the range of gemcitabine concentrations in this study was consistent with prior reports (24–26).

There were no clinically significant increases in SCN levels compared with baseline values (predose on cycle 0, day 1) following administration of either 60 mg or 80 mg of GDC-0425 (20).

Investigator assessments of best overall response were available for 31 of 40 patients with ≥1 post-baseline assessment (n = 28 patients with disease measured by RECIST v1.1); 9 patients (23%) did not have post-baseline tumor response assessments and were classified as nonresponders. A best overall response of stable disease or PR was observed in 24 (60%) patients. There were 2 confirmed PRs (5% of patients) by RECIST in patients with TNBC (TP53-mutated) and melanoma (n = 1 each) and 1 unconfirmed PR in a patient with CUP. Computed tomographic (CT) response and TP53 status by tumor type are shown in Fig. 2A.

The median duration of therapy was 3.5 cycles (range, 1–14). Eight (20%) patients remained on study treatment for more than 6 months [>8 cycles; cervical cancer, CUP, fallopian tube cancer, mesothelioma (n = 1 each); melanoma and TNBC (n = 2 each)], including 2 patients with PR (melanoma and TNBC) who received therapy for more than 10 months (Fig. 2B).

HT-29 xenografts were analyzed for pCDK1/2 and Ki-67 expression (Supplementary Fig. S2). Representative images from 2 unique animals in each treatment group (n = 5 animals per group) are shown. Ki-67 expression in surviving tumor cells is relatively consistent among treatment groups. pCDK1/2 expression increases from baseline after gemcitabine treatment, whereas GDC-0425 alone had little effect. In contrast, GDC-0425 given 16 hours after gemcitabine caused marked inhibition of the gemcitabine-induced expression of pCDK1/2.

To assess the impact of GDC-0425 and gemcitabine on Chk1-regulated cell-cycle checkpoints, pCDK1/2 was assayed in tumor tissues collected before study treatment, 24 hours after GDC-0425 administration, 48 hours after gemcitabine administration, and after the combination of GDC-0425 and gemcitabine treatment (24 hours after GDC-0425 and 48 hours after gemcitabine; Fig. 3). Tumor samples were obtained from a total of 8 patients, with 6 patients able to undergo biopsies before study treatment and after gemcitabine administration and 3 of these patients also able to undergo biopsies after treatment with the combination of GDC-0425 and gemcitabine. Representative images from tumor biopsies are shown in Supplementary Fig. S3. Increased pCDK1/2 was observed after gemcitabine administration [mean fold change, 2.01; 95% confidence interval (CI), 1.25–3.23], consistent with checkpoint activation and cell-cycle arrest in response to gemcitabine-induced DNA damage. Following GDC-0425 and gemcitabine administration, pCDK1/2 was decreased (mean fold change, 0.54; 95% CI, 0.30–0.96), consistent with checkpoint override following Chk1 inhibition.

Our findings in this first-in-human trial show that the Chk1 inhibitor GDC-0425 can be combined with a standard dose and schedule of gemcitabine when administered as a single dose approximately 24 hours after gemcitabine 1,000 mg/m² on days 1 and 8 of a 21-day cycle. Bone marrow toxicity was manageable but prevented escalating beyond 60 mg GDC-0425.

Consistent with its cell-cycle–linked mechanism of action, antitumor efficacy of a Chk1 inhibitor is predicted to potentiate the effects of DNA-damaging chemotherapy. Furthermore, preclinical models have demonstrated marked synergy when inhibiting Chk1 in combination with DNA-damaging chemotherapy that acts in S-phase, including antimetabolites such as gemcitabine (refs. 8, 18; Gazzard, manuscript in preparation). Although gemcitabine is only FDA-approved in ovarian, breast, NSCLC, and pancreatic cancers, it is known to have activity across a wide range of solid tumors pointing to the potential for broad applicability when paired with a Chk1 inhibitor.

Our findings suggest that this strategy is feasible but not without increased toxicity. Although limited toxicity was observed with brief exposure to single agent GDC-0425, chemotherapy-related toxicities may be enhanced by the addition of GDC-0425. When GDC-0425 is given in combination with gemcitabine administered at a standard dose and schedule, bone marrow toxicity is increased beyond what would be expected with gemcitabine alone. Neutropenia and thrombocytopenia were manageable but were grade 3 or 4 and treatment-related in 40% and 15% of patients, respectively, and appear to be more frequent than what would be expected with gemcitabine alone (27).

The addition of GDC-0425 also increased the rates of nonhematologic AEs that can adversely affect patient quality of life. Although generally grade 1 or 2, the rates of nausea, vomiting, fatigue, and pyrexia ranged from 40% to 48%. Although the number of patients in this trial was small and patients were treated with different dose levels of both GDC-0425 and gemcitabine, these observations suggest that long-term tolerability may be compromised in some patients. Fortunately, the rate of grade ≥ 3 transaminase elevations with the addition of GDC-0425 was similar to what would be observed with gemcitabine alone (28).

The timing of Chk1 inhibitor administration relative to gemcitabine and how this relates to maximum chemopotentiation remains unclear. Although some Chk1 inhibitors have been evaluated with concurrent chemotherapy dosing (29, 30), nonclinical studies suggest that Chk1 inhibitor administration was most effective when dosed with a defined delay after gemcitabine of approximately 24 hours (9, 31). In our phase I study, 3 days of GDC-0425 starting 24 hours after gemcitabine was not tolerable. GDC-0425 exposures in humans were greater than predicted from the animal models and the half-life of approximately 15 hours was longer than expected. Furthermore, a single dose of 60 mg in humans exceeded target exposures associated with checkpoint abrogation and antitumor activity in preclinical models. Importantly, no PK interaction was observed between GDC-0425 and gemcitabine, suggesting that the observed AEs were not due to unexpected changes in exposures with this combination. Before conducting this clinical study, a mass balance study in rats using radiolabeled GDC-0425 was performed and one of the drug-related metabolites identified in rat plasma was SCN (20). The overall contribution of drug-derived (radiolabeled) SCN to the endogenous SCN concentrations in rats was negligible. However, as part of the safety and PK assessments, the concentrations of SCN were monitored before and after dosing in this clinical study. There were no clinically significant increases in SCN levels compared with baseline values following GDC-0425 administration mitigating the concerns for any elevated SCN levels on dosing with GDC-0425.

Preliminary signs of clinical activity were observed supporting this chemopotentiation strategy. Eight of 40 patients (20%) remained on study for more than 6 months and 3 PRs were observed in patients with melanoma, TNBC, and CUP. It has been hypothesized that chemopotentiation with a Chk1 inhibitor which abrogates the S and G₂ cell-cycle checkpoints may be more effective in patients whose tumors lack functional p53. Interestingly, despite the small number of tumor biopsies available for PD biomarker analyses, 2 patients with TP53-mutant tumors showed the predicted decrease in pCDK1/2 after treatment with the combination of gemcitabine and GDC-0425, whereas a third patient with TP53 wild-type tumor did not show this effect. Overall, 21 of the 28 patients with RECIST-evaluable disease and on-treatment tumor assessment had adequate archival tumor tissue analyzed for TP53 mutation status. Although 2 of the 3 patients with a PR and others with minor radiographic changes had mutations in TP53, there are insufficient data from this dose-escalation trial to understand the correlation between TP53 status and clinical activity.

In summary, this trial provides further evidence that Chk1 inhibition is an attractive therapeutic strategy to enhance the cytotoxicity of DNA damaging chemotherapeutics. Preclinical models suggest that antimetabolites, such as gemcitabine, are a preferred partner to pair with Chk1 inhibitors to accentuate chemopotentiation. Our findings show that combining a standard dose and schedule of gemcitabine with GDC-0425 is feasible but the addition of the Chk1 inhibitor likely adds toxicity that will need to be accounted for in future trials. The exposures achieved, lack of a PK interaction, PD changes in tumor samples, and early clinical activity in this diverse solid tumor patient population is encouraging support for the further development of this chemopotentiation strategy. As future studies are conducted, it will be important to evaluate tolerability and antitumor activity of GDC-0425 and gemcitabine in a randomized fashion in a less heavily treated patient population and to understand whether tumors that lack functional p53 may afford greater opportunity to leverage chemopotentiation for clinical benefit and eventual patient selection.

E.M. Blackwood, M. Evangelista, S. Mahrus, F.V. Peale, X. Lu, S. Sahasranaman, R. Zhu, Y. Chen, X. Ding, E.R. Murray, J.L. Schutzman, and J.O. Lauchle are shareholders of F. Hoffmann La Roche, Ltd. J.-C. Soria is a consultant/advisory board member for Roche. P.M. LoRusso is a consultant/advisory board member for Genentech. No potential conflicts of interest were disclosed by the other authors.

The authors take full responsibility for the design of the study, the collection of the data, the analysis and interpretation of the data, the decision to submit the article for publication, and the writing of the article.

Conception and design: J.R. Infante, E.M. Blackwood, S. Mahrus, S. Sahasranaman, J.-C. Soria, P.M. LoRusso

Development of methodology: J.R. Infante, E.M. Blackwood, M. Evangelista, S. Mahrus, S. Sahasranaman, P.M. LoRusso

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): J.R. Infante, A. Hollebecque, S. Postel-Vinay, T.M. Bauer, J.-C. Soria, P.M. LoRusso

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): J.R. Infante, S. Mahrus, F.V. Peale, X. Lu, S. Sahasranaman, R. Zhu, Y. Chen, X. Ding, J.L. Schutzman, J.O. Lauchle, J.-C. Soria, P.M. LoRusso

Writing, review, and/or revision of the manuscript: J.R. Infante, A. Hollebecque, S. Postel-Vinay, T.M. Bauer, E.M. Blackwood, M. Evangelista, S. Mahrus, F.V. Peale, X. Lu, S. Sahasranaman, R. Zhu, Y. Chen, E.R. Murray, J.L. Schutzman, J.O. Lauchle, J.-C. Soria, P.M. LoRusso

Study supervision: J.R. Infante, E.R. Murray, J.L. Schutzman, J.O. Lauchle, P.M. LoRusso

Other (study conduction): S. Postel-Vinay

The authors wish many thanks to all of the patients and the investigators who participated in this study. We thank Shari Lau for IHC support. Writing assistance was provided by Genentech, Inc.

This work was supported by Genentech, Inc.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.