Purpose:

Targeting focal adhesion kinase (FAK) renders checkpoint immunotherapy effective in pancreatic ductal adenocarcinoma (PDAC) mouse model. Defactinib is a highly potent oral FAK inhibitor that has a tolerable safety profile.

Patients and Methods:

We conducted a multicenter, open-label, phase I study with dose escalation and expansion phases. In dose escalation, patients with refractory solid tumors were treated at five escalating dose levels of defactinib and gemcitabine to identify a recommended phase II dose (RP2D). In expansion phase, patients with metastatic PDAC who progressed on frontline treatment (refractory cohort) or had stable disease (SD) after at least 4 months of standard gemcitabine/nab-paclitaxel (maintenance cohort) were treated at RP2D. Pre- and posttreatment tumor biopsies were performed to evaluate tumor immunity.

Results:

The triple drug combination was well-tolerated, with no dose-limiting toxicities. Among 20 treated patients with refractory PDAC, the disease control rate (DCR) was 80%, with one partial response (PR) and 15 SDs, and the median progression-free survival (PFS) and overall survival (OS) were 3.6 and 7.8 months, respectively. Among 10 evaluable patients in the maintenance cohort, DCR was 70% with one PR and six SDs. Three patients with SD came off study due to treatment- or disease-related complications. The median PFS and OS on study treatment were 5.0 and 8.3 months, respectively.

Conclusions:

The combination of defactinib, pembrolizumab, and gemcitabine was well-tolerated and safe, had promising preliminary efficacy, and showed biomarker activity in infiltrative T lymphocytes. Efficacy of this strategy may require incorporation of more potent chemotherapy in future studies.

Translational Relevance

Checkpoint immunotherapy is now an integral treatment option for many cancer types. However, the efficacy of this modality is limited in PDAC due to the various biological obstacles, particularly the immune-suppressive tumor microenvironment that consists of a dense stroma and heavy infiltration of suppressive myeloid cells. Preclinical PDAC model showed that targeting focal adhesion kinase (FAK) could overcome these obstacles and induce responsiveness to anti–PD-1, especially when combined with chemotherapy. On this premise, we conducted a phase I clinical study combining a FAK inhibitor (defactinib), anti–PD-1 (pembrolizumab), and gemcitabine in patients with PDAC. Our results showed that this approach was safe and well-tolerated, and resulted in promising antitumor activity in patients with PDAC in both refractory and maintenance settings. Analysis of tumor biopsies showed reduced suppressive macrophages and increased infiltration of cytotoxic T cells. A randomized controlled clinical trial that includes more potent chemotherapy backbone and biomarker studies may be considered.

Pancreatic ductal adenocarcinoma (PDAC) remains highly lethal with no effective systemic treatment options other than combination chemotherapies, which are palliative, typically with short-lived treatment response and with considerable toxicities (1, 2). The benefit of immune or targeted therapies is limited to a small subset of patients with known tumor-specific or germline mutations. For instance, checkpoint immunotherapies, including antagonists against programed cell dealth-1 (PD-1) or cytotoxic T-lymphocyte-associated protein 4 are only beneficial to ∼0.8% of patients with mismatch repair-deficient (dMMR) or microsatellite stability (MSS) disease (3–5). PARP inhibitors such as olaparib serves a nonchemotherapy option for patients with PDAC with germline breast cancer gene 1 (BRCA-1) and BRCA-2 mutations after disease stabilization with platinum-based chemotherapy such as FOLFIRINOX (cocktail of folinic acid, irinotecan, oxaliplatin, and 5-fluorouracil; refs. 6, 7). Therefore, novel therapeutic strategies that can enhance and prolong treatment response with minimal side effects are direly needed.

Focal adhesion kinase (FAK) is a tyrosine kinase regulating multiple cellular functions, in particular cellular organization, such as adhesion and migration (8, 9). FAK is hyperactivated in the majority of PDAC, and its level of activation is associated with poor prognosis (10). In preclinical PDAC mouse models, pharmacologic FAK inhibition results in delayed tumor growth and better survival, and the antitumor effect is boosted when combined with anti-PD1 and gemcitabine (10). Mechanistically, FAK inhibition resulted in decreased tumor desmoplasia and tumor-associated macrophages (TAM) and increased influx of cytotoxic T-cell infiltration (10, 11). Similarly, FAK's immune-suppressive effects on the tumor microenvironment have been reported in other tumor types, including squamous cell carcinoma (8). Defactinib (vs. -6063, Verastem) is a potent and safe, orally administered small molecule FAK inhibitor which directly competes for the ATP-binding site (12–14). Here, we present the safety, preliminary efficacy, and biomarker data for the triple drug combination of defactinib, pembrolizumab, and gemcitabine in advanced solid tumors, focusing on PDAC.

Study design and participants

We conducted an open-label, phase I study, which included a dose escalation and a dose expansion at Washington University in St. Louis and Mayo Clinic in Rochester (Supplementary Tables S1 and S2), to determine the safety of a combination of defactinib, pembrolizumab, and gemcitabine in patients (≥18 years of age) with histologically confirmed advanced solid tumors (dose escalation cohort) or advanced PDAC (dose expansion cohort). The one dose expansion cohort (refractory cohort) included patients with advanced PDAC, who progressed with at least one line of systematic therapy for advanced disease. The second expansion cohort (maintenance cohort) enrolled patients with PDAC who achieved a treatment response or stable disease after at least 4 months of frontline therapy of gemcitabine and Abraxane (G/A). Toxicity assessment was performed on all treated patients, regardless of evaluability of treatment response. Treated patients who received imaging scans prior to the first planned assessment (after three cycles) due to clinical suspicion of progression were deemed evaluable. Patients who came of study prior to any imaging evaluation to document disease status were included in survival analyses if deemed to be related to clinical disease progression. Measurable disease, defined as lesions ≥10 mm with a CT scan or MRI, ≥20 mm by chest X-rays, or ≥10 mm with calipers by clinical exam, was required. In addition, patients were deemed eligible if they had an Eastern Cooperative Oncology Group performance status of ≤1 and a life expectancy of at least 3 months. Patients who had received prior FAK inhibitors or anti–PD-1, anti-PD-L1, anti-PD-L2, anti-CD137, or anti-cytotoxic T-lymphocyte-associated antigen 4 antibody were ineligible, as were patients with brain metastasis. The study was performed in accordance with the Declaration of Helsinki statement on ethical biomedical research and with the International Conference on Harmonization Guidelines for Good Clinical Practice. The study was approved by the local Institutional Review Boards for each study site and registered at clinical trial.gov as NCT02546531. All patients provided written informed consent. Patient cohort demographic data are in Supplementary Tables S1 and S2.

Procedures

Dose escalation was conducted using a two-stage design. A small cohort of 3 patients were treated at the initial dose level. If no DLT was observed, the next cohort was assigned to a higher dose level. As soon as the first DLT is observed, three additional patients were enrolled. For the dose escalation cohort, defactinib was started at twice per day at either a 200 mg twice daily or 400 mg twice daily on a 21-day cycle. Pembrolizumab (200 mg) was given intravenously every 21 days. Gemcitabine was not given to the first two dose levels and was administered on days 1 and 8 of a 21-day cycle at 500, 750, or 1,000 mg/m2 for dose levels 3, 4, or 5, respectively (Fig. 1). The recommended phase II dose (RP2D) was determined to be defactinib 400 mg twice daily D1–21, pembrolizumab 200 mg D1, and gemcitabine 1,000 mg/m2 D1 and D8, every 21 days. Toxicities were assessed according to CTCAE 4.0, and the treatment response was evaluated every three cycles per RECIST v1.1.

Figure 1.

CONSORT flow diagram of patients enrolled in this study.

Figure 1.

CONSORT flow diagram of patients enrolled in this study.

Close modal

Clinical outcomes measures

The primary endpoint was to assess the safety and toxicity profile of the triple-drug regimen and determine the MTD or RP2D. The secondary endpoints included objective response rate (ORR), time on treatment, PFS and OS, which were described using Kaplan–Meier product limit method. The exploratory endpoints were to assess biomarkers associated with treatment response and resistance.

Multiparametric flow cytometry of biopsy tissues

Biopsy tissues were collected from the endoscopy suite and placed in DMEM on wet ice prior to processing. Human tumor samples were digested in HBSS supplemented with 2 mg/mL collagenase A (Roche), 2.5 U/mL hyaluronidase, and DNase I at 37°C for 30 minutes with agitation to generate single cell suspensions. Following tissue digestion, single cell suspensions were resuspended in flow cytometry buffer (PBS containing 1% BSA and 5 mmol/L EDTA), FcR blocked with human TruStain FcX (BioLegend) for 10 minutes, and pelleted by centrifugation. When applicable, live/dead viability dyes were used for 15 to 30 minutes at room temperature prior to centrifugation. The cells were consequently labeled with 200 μL of fluorophore-conjugated anti-human extracellular antibodies at recommended dilutions for 30 minutes on ice. Intracellular staining for transcription factors and intracellular markers was subsequently employed using the eBioscience Transcription Factor Staining Buffer Set (eBioscience), according to the manufacturer's instructions. Flow cytometry data were acquired using BD Fortessa X-20 (BD Biosciences) within 7 days and analyzed using FlowJo software (version 10) with application of bead-based post hoc compensation.

Statistical analysis

All statistical analyses were performed using GraphPad Prism software (GraphPad) or by a biostatistics core expert at Washington University using SAS 9.4 (SAS Institutes). Data are shown as the mean ± SEM, unless otherwise noted. Normal distribution of data was assessed using the D'Agostino-Pearson omnibus normality test in Prism (GraphPad). For survival analyses, the log-rank (Mantel–Cox) test was used. For proximity analyses, the nonparametric Kolmogorov–Smirnov test was used to distinguish differences in frequency distributions.

Data availability

The data generated in this study are available upon request from the corresponding author.

In total 59 patients were screed and 42 eligible patients were enrolled and treated in two phases: dose escalation (n = 20) and expansion phase (n = 22, Fig. 1). Of the 20 patients treated in the dose escalation, 12 (60%) had refractory PDAC who had progressed on frontline chemotherapy. The median age was 62 years, and the median number of prior systemic treatments was two lines. All 22 patients enrolled in the dose expansion phase had PDAC. In this portion, 11 patients had PDAC which had progressed on one to three different combination chemotherapies (refractory cohort), and 11 with stable disease after at least 4 months of frontline Gemcitabine plus Abraxane (or nab-paclitaxel, G/A, maintenance cohort). None of the patients with tumor tested with next-generation sequencing (NGS) had were microsatellite unstable or high tumor mutational burden (TMB) disease (Table 1).

Table 1.

Baseline patient demographics and disease characteristics.

All patientsDose escalation (n = 20)Dose expansion (n = 22)
Age, years 62 (46–75) 64 (35–84) 
Sex 
 Female 13 (65%) 12 (54%) 
 Male 7 (35%) 10 (46%) 
Race 
 White 16 (80%) 20 (91%) 
 Black 3 (15%) 
 Others 1 (5%) 2 (9%) 
ECOG PS 
 0 5 (25%) 3 (14%) 
 1 15 (75%) 19 (86%) 
Histology 
 PDAC 12 (60%) 22 (100%) 
 Biliary 3 (15%)  
 Others 5 (25%)  
PDAC patients PDAC patients (n = 12) PDAC patients (n = 22) 
  Refractory cohort (n = 11) 
  Maintenance cohort (n = 11) 
Adenocarcinoma 12 (100%) 22 (100%) 
Median prior lines of therapy (range) 2 (1–7) Refractory cohort: 2 (1–3) Maintenance cohort: 1 
Pretreatment labs 
 Albumin (>3.5 g/dL) 12 (100%) 22 (100%) 
 CA 19-9 (>35 U/mL) 9 (75%) 18 (82%) 
Prior surgery for PDAC 4 (20%) 1 (4.5%) 
Extent of disease 
 Clinical stage IV 20 (100%) 22 (100%) 
 Lung 6 (50%) 8 (37%) 
 Liver 6 (50%) 15 (68%) 
 Lymph nodes 2 (16%) 4 18%) 
 Bone 1 (8%) 2 (9%) 
 Muscle 2 (16%) 1 (4.5%) 
 Peritoneal cavity 1 (8%) 3 (14%) 
Tumor NGS 
 Available 6 (50%) 15 (68%) 
 MSS/dMMR 0 (0%) 0 (0%) 
 TMB-high 0 (0%) 0 (0%) 
All patientsDose escalation (n = 20)Dose expansion (n = 22)
Age, years 62 (46–75) 64 (35–84) 
Sex 
 Female 13 (65%) 12 (54%) 
 Male 7 (35%) 10 (46%) 
Race 
 White 16 (80%) 20 (91%) 
 Black 3 (15%) 
 Others 1 (5%) 2 (9%) 
ECOG PS 
 0 5 (25%) 3 (14%) 
 1 15 (75%) 19 (86%) 
Histology 
 PDAC 12 (60%) 22 (100%) 
 Biliary 3 (15%)  
 Others 5 (25%)  
PDAC patients PDAC patients (n = 12) PDAC patients (n = 22) 
  Refractory cohort (n = 11) 
  Maintenance cohort (n = 11) 
Adenocarcinoma 12 (100%) 22 (100%) 
Median prior lines of therapy (range) 2 (1–7) Refractory cohort: 2 (1–3) Maintenance cohort: 1 
Pretreatment labs 
 Albumin (>3.5 g/dL) 12 (100%) 22 (100%) 
 CA 19-9 (>35 U/mL) 9 (75%) 18 (82%) 
Prior surgery for PDAC 4 (20%) 1 (4.5%) 
Extent of disease 
 Clinical stage IV 20 (100%) 22 (100%) 
 Lung 6 (50%) 8 (37%) 
 Liver 6 (50%) 15 (68%) 
 Lymph nodes 2 (16%) 4 18%) 
 Bone 1 (8%) 2 (9%) 
 Muscle 2 (16%) 1 (4.5%) 
 Peritoneal cavity 1 (8%) 3 (14%) 
Tumor NGS 
 Available 6 (50%) 15 (68%) 
 MSS/dMMR 0 (0%) 0 (0%) 
 TMB-high 0 (0%) 0 (0%) 

Toxicity assessment

In the dose escalation phase, all five dose levels (DL) were well-tolerated by study participants. No dose-limiting toxicity (DLT) or treatment-related death was observed (Table 2). The recommended phase two dose (RP2D) was defactinib 400 mg twice daily, pembrolizumab 200 mg on day 1, and gemcitabine 1,000 mg/m2 on days 1 and 8 of a 3-week cycle. The most common treatment-related adverse events (TRAE) were fatigue (65%), anorexia (43%), neutropenia (57%), anemia (57%), thrombocytopenia (35%), nausea (52%), vomiting (39%), and edema (30%). Grade 3 or 4 TRAEs occurring in more than 10% of patients included neutropenia (39%) and anemia (17%). Immune-related AEs included one case of mild, self-limited pancreatitis, which was attributed possibly to defactinib or pembrolizumab, one case of mild hypothyroidism attributed to pembrolizumab, and four cases of grade 1 to 2 dyspnea secondary to pneumonitis which were attributed to all agents. All these 4 patients recovered after temporary treatment break without requiring the use of steroids, and none had recurrence after treatment rechallenge.

Table 2.

Treatment-related AEs (including >10% for all grades, and all grade 3/4).

 Dose escalation (n = 20)
DL1, n = 3DL2, n = 3DL3, n = 3DL4, n = 7DL5 (RP2D), n = 4Dose expansion (RP2D), n = 22
 G 3/4AllG 3/4AllG 3/4AllG 3/4AllG 3/4AllG3/4All
Constitutional             
 Anorexia 2 (67%)p,d 1 (33%)p,d,g 1 (14%)g 1 (25%)g 10 (43%)p,d,g 
 Weakness 1 (33%)p,d 1 (33%)p,d 
 Fatigue 1 (33%)p,d 2 (67%)p,d 1 (33%)p,d 2 (67%)p,d,g 2 (29%)g 1 (25%)d 3 (75%)p,d,g 1 (4%)p,d,g 15 (65%)p,d,g 
 Edema 1 (25%)g 7 (30%)p,d,g 
 Fever 1 (33%)g 2 (29%)d,g 4 (17%)p,d,g 
 Myalgia 1 (33%)p,d 1 (33%)g 2 (9%)p 
 Arthralgia 1 (33%)p 1 (33%)p 1 (14%)p,g 1 (25%)d 1 (4%)p 
Hematologic             
 Neutropenia 1 (14%)g 2 (29%)g 1 (25%)p,d,g 1 (25%)p,d,g 9 (39%)p,d,g 13 (57%)p,d,g 
 Leukopenia 1 (33%)d,g 1 (33%)d,g 4 (57%)g 2 (50%)g 6 (26%)p,d,g 
 Anemia 1 (33%)g 3 (43%)g 3 (75%)g 4 (17%)p,d,g 13 (57%)p,d,g 
 Thrombocytopenia 1 (33%)g 1 (33%)g 2 (29%)g 1 (4%)g 8 (35%)p,d,g 
Gastrointestinal             
 Nausea 2 (67%)p,d 1 (33%)p 2 (67%)p,d,g 2 (29%)g 1 (25%)g 12 (52%)p,d,g 
 Vomiting 2 (67%)p,d 1 (33%)p,d,g 2 (29%)p,d,g 1 (25%)g 2 (9%)p,d,g 9 (39%)p,d,g 
 Diarrhea 1 (33%)p,d 1 (33%)p 1 (4%)g 8 (35%)p,d,g 
 Constipation 1 (33%)p,d 1 (4%)p,d,g 
 Abdominal pain 1 (4%)p,d,g 
 Mucositis 1 (33%)p 1 (14%)g 2 (9%)p,d,g 
 Pancreatitis 1 (4%)d 
 ALT/AST increased 1 (14%)d,g 1 (14%)d,g 2 (9%)p,d,g 
 Elevated bilirubin 1 (25%)p,g 1 (4%)p,g 1 (4%)p,g 
Renal and electrolytes             
 Creatinine increased 1 (25%)p,g 1 (25%)p,g 
 Hypokalemia 1 (4%)g 1 (4%)g 
 Hypernatremia 1 (4%)g 
Cardiovascular             
 Hypertension 2 (9%)p,g 2 (9%)p,g 
 Heart failure 1 (4%)p,g 2 (9%)p,g 
 DVT 1 (4%)p 1 (4%)p 
Pulmonary             
 Pneumonitis with dyspnea 4 (17%)p,d,g 
 Hypoxia 1 (4%)p,d,g 
Endocrine             
 Hypothyroidism 1 (33%)p 
Dermatologic             
 Alopecia 4 (17%)p,d,g 
 Rash 1 (25%)d 6 (26%)p,d,g 
 Dry skin 1 (4%)p,d,g 
 Pruritus 1 (33%)d 2 (29%)p,g 1 (4%)d 
 Dose escalation (n = 20)
DL1, n = 3DL2, n = 3DL3, n = 3DL4, n = 7DL5 (RP2D), n = 4Dose expansion (RP2D), n = 22
 G 3/4AllG 3/4AllG 3/4AllG 3/4AllG 3/4AllG3/4All
Constitutional             
 Anorexia 2 (67%)p,d 1 (33%)p,d,g 1 (14%)g 1 (25%)g 10 (43%)p,d,g 
 Weakness 1 (33%)p,d 1 (33%)p,d 
 Fatigue 1 (33%)p,d 2 (67%)p,d 1 (33%)p,d 2 (67%)p,d,g 2 (29%)g 1 (25%)d 3 (75%)p,d,g 1 (4%)p,d,g 15 (65%)p,d,g 
 Edema 1 (25%)g 7 (30%)p,d,g 
 Fever 1 (33%)g 2 (29%)d,g 4 (17%)p,d,g 
 Myalgia 1 (33%)p,d 1 (33%)g 2 (9%)p 
 Arthralgia 1 (33%)p 1 (33%)p 1 (14%)p,g 1 (25%)d 1 (4%)p 
Hematologic             
 Neutropenia 1 (14%)g 2 (29%)g 1 (25%)p,d,g 1 (25%)p,d,g 9 (39%)p,d,g 13 (57%)p,d,g 
 Leukopenia 1 (33%)d,g 1 (33%)d,g 4 (57%)g 2 (50%)g 6 (26%)p,d,g 
 Anemia 1 (33%)g 3 (43%)g 3 (75%)g 4 (17%)p,d,g 13 (57%)p,d,g 
 Thrombocytopenia 1 (33%)g 1 (33%)g 2 (29%)g 1 (4%)g 8 (35%)p,d,g 
Gastrointestinal             
 Nausea 2 (67%)p,d 1 (33%)p 2 (67%)p,d,g 2 (29%)g 1 (25%)g 12 (52%)p,d,g 
 Vomiting 2 (67%)p,d 1 (33%)p,d,g 2 (29%)p,d,g 1 (25%)g 2 (9%)p,d,g 9 (39%)p,d,g 
 Diarrhea 1 (33%)p,d 1 (33%)p 1 (4%)g 8 (35%)p,d,g 
 Constipation 1 (33%)p,d 1 (4%)p,d,g 
 Abdominal pain 1 (4%)p,d,g 
 Mucositis 1 (33%)p 1 (14%)g 2 (9%)p,d,g 
 Pancreatitis 1 (4%)d 
 ALT/AST increased 1 (14%)d,g 1 (14%)d,g 2 (9%)p,d,g 
 Elevated bilirubin 1 (25%)p,g 1 (4%)p,g 1 (4%)p,g 
Renal and electrolytes             
 Creatinine increased 1 (25%)p,g 1 (25%)p,g 
 Hypokalemia 1 (4%)g 1 (4%)g 
 Hypernatremia 1 (4%)g 
Cardiovascular             
 Hypertension 2 (9%)p,g 2 (9%)p,g 
 Heart failure 1 (4%)p,g 2 (9%)p,g 
 DVT 1 (4%)p 1 (4%)p 
Pulmonary             
 Pneumonitis with dyspnea 4 (17%)p,d,g 
 Hypoxia 1 (4%)p,d,g 
Endocrine             
 Hypothyroidism 1 (33%)p 
Dermatologic             
 Alopecia 4 (17%)p,d,g 
 Rash 1 (25%)d 6 (26%)p,d,g 
 Dry skin 1 (4%)p,d,g 
 Pruritus 1 (33%)d 2 (29%)p,g 1 (4%)d 

Note: Attributed to ppembrolizumab, ddefactinib, or ggemcitabine.

Preliminary efficacy

Of 12 patients with PDAC enrolled in dose escalation, 2 patients withdrew during cycle 1 to pursue hospice (DL4, lost to follow up) or gemcitabine monotherapy (DL4, CT later showed SD) and both were excluded from efficacy evaluation. Two patients were removed by treating physicians due to severe fatigue attributed to TRAE (DL5, <1 cycle, CT showed SD) or underlying disease (DL5, < 2 cycles, CT showed PD) and were deemed evaluable. Eight patients underwent planned radiographic assessment after three cycles. Of all 10 evaluable patients, 1 patient experienced a partial response (PR), 7 had stable disease (SD) and 2 had PD, resulting in a disease control rate (DCR, equals PR + SD) of 80% (8 of 10). For the 8 patients who underwent planned radiographic assessment, DCR was 87.5% (6 SD, 1 PR, 1 PD, Fig. 2AD). Notably, two patients who previously progressed on gemcitabine stayed on study for more than one year (case #01–005: DL2 stayed on study for >24 months, Fig. 3A; case #01–026: DL5 with PR stayed on study for 12 months, Fig. 3B). Another patient (case #01–028: DL5) who was gemcitabine-naive stayed on study for 12.4 months (Fig. 2A). NGS showed all 3 patients had MSS or TMB-low disease.

Figure 2.

A, Swimmer plot depicting prior treatment history and time on study for all patients with PDAC enrolled in this study. Depicted on the plot are symbols for treatment response during study treatment and subsequent treatment (for maintenance cohort) after coming off study. Patient 1: Off study per physician discretion due to severe fatigue and opted for hospice. Patient 2: Developed congestive heart failure deemed unrelated to gemcitabine, defactinib, and pembrolizumab. Patient 3: Whipple procedure attempted and aborted due to intraoperative finding of liver metastases and later died from postoperative wound infection. Patient 4: Poor oral intake despite stable CT scans and pursued hospice. B, Waterfall plot depicting best radiographic response for patients treated in each cohort. Patients who had imaging studies and came off study prior to planned (after 3 cycles) evaluation per RECIST were included. C, Kaplan–Meier curves depicting the PFS and OS for refractory patients with PDAC from dose escalation (n = 10) and refractory expansion (n = 10). D, Kaplan–Meier curves depicting the PFS and OS for patients with PDAC treated in the dose expansion phase as maintenance cohort (n = 10).

Figure 2.

A, Swimmer plot depicting prior treatment history and time on study for all patients with PDAC enrolled in this study. Depicted on the plot are symbols for treatment response during study treatment and subsequent treatment (for maintenance cohort) after coming off study. Patient 1: Off study per physician discretion due to severe fatigue and opted for hospice. Patient 2: Developed congestive heart failure deemed unrelated to gemcitabine, defactinib, and pembrolizumab. Patient 3: Whipple procedure attempted and aborted due to intraoperative finding of liver metastases and later died from postoperative wound infection. Patient 4: Poor oral intake despite stable CT scans and pursued hospice. B, Waterfall plot depicting best radiographic response for patients treated in each cohort. Patients who had imaging studies and came off study prior to planned (after 3 cycles) evaluation per RECIST were included. C, Kaplan–Meier curves depicting the PFS and OS for refractory patients with PDAC from dose escalation (n = 10) and refractory expansion (n = 10). D, Kaplan–Meier curves depicting the PFS and OS for patients with PDAC treated in the dose expansion phase as maintenance cohort (n = 10).

Close modal
Figure 3.

Characteristics of 2 patients with partial response. A and B, Prior treatment history, molecular features of tumors, serum CA19–9, and serial radiographic images of 2 patients who had radiographic partial response on study.

Figure 3.

Characteristics of 2 patients with partial response. A and B, Prior treatment history, molecular features of tumors, serum CA19–9, and serial radiographic images of 2 patients who had radiographic partial response on study.

Close modal

In the dose expansion refractory cohort (n = 11), 1 patient withdrew to pursue hospice before treatment start and was excluded from efficacy evaluation. Of the remaining 10 patients, 6 had not been treated with a gemcitabine-based regimen before enrollment into our study. Two patients were removed due to worsening symptoms during cycle 2 and imaging showed PD. The remaining 8 patients who underwent planned radiographic assessment after three cycles per protocol all had SD. Of all these 10 patients, 2 had PD and 8 had SD, leading to an initial DCR of 80% (8 of 10 treated patients). Because patients with PDAC treated at dose escalation phase (n = 10) and refractory expansion phase (n = 10) both had refractory metastatic PDAC progressed on at least one line of chemotherapy, we combined the efficacy data from these two cohorts to obtain a clearer picture on the benefit of defactinib plus pembrolizumab. Of these 20 patients with refractory disease, the first planned or unplanned imaging showed 4 PD, 1 PR, 15 SD, achieving an initial DCR of 80% (Fig. 2B). The median duration of study treatment was 3.2 months (range, 0.7–12.6 months, Fig. 2C). The median progression-free survival (PFS) was 3.6 months (90% CI, 2.9–6.2 months), and the median overall survival (OS) was 7.8 months (90% CI, 5.7–10.3 months).

In the dose expansion maintenance cohort, 11 patients who had SD after at least 4 months of frontline Gemcitabine plus Abraxane (G/A) were enrolled. One patient developed mental confusion on cycle 1 day 1 of study, probably due to premedication including dexamethasone, and subsequently decided to pursue hospice after recovery. This patient was deemed not efficacy-evaluable. Of the remaining 10 evaluable patients, 1 had a delayed PR, 6 continued to have SD, and 3 had PD, achieving a DCR of 70% (7 of 10 treated patients, Fig. 2A and B). The patient (case #01–0145) with delayed PR stayed on treatment for 16.1 months and had MSS and TMB-low tumor (Fig. 3B). Three patients with SD came off study and passed away shortly after due to treatment or disease-related complications including a patient (case #01–046) who developed congestive heart failure deemed unrelated to gemcitabine, defactinib, and pembrolizumab; another patient (case #02–007) who struggled with poor oral intake and decided to pursue hospice; and the third patient (case #01–044) underwent Whipple procedure which was aborted due to intraoperative findings of small liver metastases and later died from poor wound healing. The remaining four patients who later progressed on study treatment went on to receive other treatment. The median duration of treatment was 5.0 months (range 1.6–15.8 months, Fig. 2A). The median PFS and OS since initiation of study treatment were 5.0 months (90% CI, 2.8–14.2 months) and 8.3 months (90% CI, 4.2–21.3 months), respectively (Fig. 2D). For patients in maintenance cohort, median PFS and OS from initial treatment with G/A was 11.3 months (90% CI, 7.9–21.7 months) and 20.6 months (90% CI, 12.1–28.4 months), respectively.

Correlative studies

Patients with PDAC in both cohorts underwent needle biopsies of tumor tissues before and after three cycles of therapy, and tumor samples were processed for flow cytometric and IHC analysis. Of all collected tissues, nine paired biopsies produced quality flow cytometry data and IHC/pathology data adequate for meaningful analysis. IHC analysis showed phospho-FAK1 (tyrosine 397), a marker of FAK auto-activation, was decreased in CK19+ PDAC cells in the majority patients biopsies tissues (Fig. 4A and B). Assessment of immune cell infiltration was evaluated by flow cytometry and IHC of biopsy tissues. Consistent with preclinical models (10), most posttreatment samples displayed increases in total CD3+CD8+ cytotoxic T lymphocytes (CTL), CD3+CD4+ FOXP3 effector T cells, and pro-Ki67+ CD8+ CTLs. In contrast, most patients showed decreased numbers of CD14+CD68+MHCIIHi TAM and CD3+CD4+ FOXP3+ regulatory T (Treg) cells. Increases in CD8+ CTLs were verified by IHC, with more marked trends observed in the number of T cells within 100 μm of CK19+ cells (Fig. 4C and D). Notably, in this small cohort, no pattern was established between patient outcomes, such as PFS or OS and changes in tumor immunity. In addition, posttreatment biopsies were not possible in two PR patients due to a decrease in the previously easily accessible metastatic lesions. Together, these data suggested that the treatment combination impacted tumor infiltration by immune cells.

Figure 4.

Correlative studies on pre- and post-biopsy tissues. A, Example image of standard IHC analysis for phosphorylated FAK1 (tyrosine 395) from paired patient samples taken pre- and posttreatment. B, Example image of multiplex IHC (left) and quantitation of pFAK high CK19+ cells in paired biopsies. C, Patient biopsies were assessed by using flow cytometry (left) or IHC analysis (right) for the presence of different leukocyte subsets. These leukocytes for flow cytometry included the following T-cell subsets: total CD8+ cytotoxic lymphocytes (CTL), Ki67+CD8+ CTLs, CD4+ T-effector cells, FOXP3+ regulatory T cells, and CD3/CD19CD11b+CD14+CD68+ TMAs. For IHC analysis, total CD8+ T cells in the biopsy and CD8+ T cells present within 100 μm of CK19+ cells were analyzed by HALO software. All data are displayed as log2fold changes between pre- and post-biopsies. D, Example IHC analyses for CD8 (brown) and CK19 (red) images for a pair of pre- and post-biopsies from a single patient. *, P < 0.05 by paired parametric test (B) or Wilcoxon signed-rank test or one-sample test (C) as appropriate.

Figure 4.

Correlative studies on pre- and post-biopsy tissues. A, Example image of standard IHC analysis for phosphorylated FAK1 (tyrosine 395) from paired patient samples taken pre- and posttreatment. B, Example image of multiplex IHC (left) and quantitation of pFAK high CK19+ cells in paired biopsies. C, Patient biopsies were assessed by using flow cytometry (left) or IHC analysis (right) for the presence of different leukocyte subsets. These leukocytes for flow cytometry included the following T-cell subsets: total CD8+ cytotoxic lymphocytes (CTL), Ki67+CD8+ CTLs, CD4+ T-effector cells, FOXP3+ regulatory T cells, and CD3/CD19CD11b+CD14+CD68+ TMAs. For IHC analysis, total CD8+ T cells in the biopsy and CD8+ T cells present within 100 μm of CK19+ cells were analyzed by HALO software. All data are displayed as log2fold changes between pre- and post-biopsies. D, Example IHC analyses for CD8 (brown) and CK19 (red) images for a pair of pre- and post-biopsies from a single patient. *, P < 0.05 by paired parametric test (B) or Wilcoxon signed-rank test or one-sample test (C) as appropriate.

Close modal

To our knowledge, this is the first clinical study evaluating an FAK inhibitor in combination with a checkpoint inhibitor and chemotherapy. Overall, the toxicities of defactinib and pembrolizumab with or without gemcitabine were very modest, with the majority of AEs being grade 1 or 2. Grade 3 AEs such as neutropenia and anemia were consistent with known side effects of gemcitabine. We did not observe significant immune-related toxicities.

The preliminary efficacy from this small study was overall encouraging, with room for improvement. For patients with PDAC who have progressed on frontline FOLFIRINOX, G/A or gemcitabine monotherapy are available options. In this setting, a retrospective study showed that gemcitabine monotherapy achieved a DCR of 35% (with 11% ORR) and median PFS of 2.5 months (15). Another retrospective study similarly showed that gemcitabine monotherapy achieved a DCR of 40%, median PFS and OS of 2.1 and 3.7 months, respectively (16). Our refractory cohort treated with the defactinib and pembrolizumab with or without gemcitabine achieved a DCR of 80%, median PFS and OS of 3.6 and 7.8 months. These results were superior to gemcitabine alone and potentially comparable with gemcitabine and nab-paclitaxel in second-line setting, which reported a DCR of 58%, median PFS and OS of 5.1 and 8.8 months respectively (17). Again, the relatively tolerable triple regimen may be an option for patients who sustained significant neuropathy from frontline FOLFIRINOX and are not candidates for nab-paclitaxel.

In the maintenance cohort, patients with PDAC who had SD after at least 4 months of first-line G/A were enrolled. For treatment naïve patients, the phase III MPACT study showed that G/A achieved a DCR of 48% (23% ORR) and median PFS and OS of 5.5 and 8.5 months, respectively (2). Patients who were treated till PD had a median OS of 9.8 months and among these >50% received subsequent therapy (18). In our study, after initial SD with at least 4 months of G/A, the triple drug regimen delivered a DCR of 63.6%, median PFS and OS of 5.0 and 8.3 months. Median PFS and OS from initiation of G/A were 11.3 and 20.6 months, respectively, as compared with 5.5 and 8.5 months in the MPACT study. It is worth also mentioning that 3 patients in the maintenance cohort dropped out without radiographic progression and passed away shortly, which could impact our survival analyses in this small study. In addition, it is entirely possible that patients who had SD after 4 months of G/A may have more favorable tumor biology. However, the triple drug regimen could potentially be more tolerable as it is not neurotoxic and less myelosuppressive then continuation of G/A. This advantage may increase the likelihood of patients being treated and tolerate subsequent treatment regimens such as mFOLFIRINOX or 5-FU/liposomal-irinotecan, which is critical in prolonging their survival. From an immunologic aspect, the defactinib/pembrolizumab/gemcitabine combination revealed the enhanced recruitment of CD8+ T cells with reduced TAM and Treg cells. However, the changes in the tumor microenvironment did not directly correlate with treatment response, which may be related to the small sample size and tumor heterogeneity. Notably, collection of posttreatment biopsies in 2 PR patients was not possible due to a decrease in the previously easily accessible metastatic lesions.

A remarkable observation from our study is prolonged SD of >4 months in a considerable number of patients (10 of 20 in refractory cohort, 6 of 10 patients in maintenance cohort, Fig. 2A). Given the well-tolerated safety profile of the triple regimen, it is reasonable to speculate that addition of nab-paclitaxel may achieve better efficacy, which can be tested in frontline setting for treatment naïve patients. Furthermore, targeting FAK was shown to synergize with radiation in preclinical PDAC models (19), and is now being evaluation in a clinical trial (NTC04331041). Although not evaluated here, preclinical studies have suggested that the relative contribution of SMAD versus STAT3 signaling may control responsiveness to FAK inhibition in PDAC mouse models (20). This may also suggest differences in tumor subtypes, such as classical and basal or underlying genomic biology (e.g., SMAD mutations), which might influence the effects of FAK inhibition. Further, patient selection criterion might benefit from selection of pFAK high patients of patients with loss of Merlin deficiency (21). But these remain to be explored.

In short, our study showed that defactinib in combination with gemcitabine and pembrolizumab is safe, well-tolerated, and showed encouraging preliminary benefit, further testing with more potent chemotherapy backbone such as G/A in a randomized controlled trial may be considered in the future.

Commercial, private foundation, or federal funding sources had no role in the study execution, data collection, analysis, or manuscript preparation.

A. Wang-Gillam reports grants from BJC Foundation during the conduct of the study as well as employment with Jacobio. R. McWilliams reports personal fees from Zentalis Pharmaceuticals outside the submitted work. H.H. Yoon reports other support from BeiGene, ALX Oncology, AstraZeneca, Bristol-Myers Squibb, Macrogenics, Merck, OncXerna, Zymeworks, Novartis, Astellas, and Amgen and grants from BeiGene, Bristol-Myers Squibb, CARsgen Therapeutics, and Merck outside the submitted work. W.G. Hawkins reports grants from NCI during the conduct of the study as well as grants from Celldex outside the submitted work. D.G. DeNardo reports grants from Verastem and NIH/NCI during the conduct of the study as well as grants from Pfizer, BMS, NIH/NCI, and PANCAN outside the submitted work. No disclosures were reported by the other authors.

A. Wang-Gillam: Conceptualization, resources, supervision, investigation, writing–original draft, writing–review and editing. K.-H. Lim: Data curation, formal analysis, supervision, investigation, writing–original draft, writing–review and editing. R. McWilliams: Investigation. R. Suresh: Investigation. A.C. Lockhart: Investigation. A. Brown: Data curation, investigation. M. Breden: Data curation, formal analysis. J.I. Belle: Investigation. J. Herndon: Investigation. S.J. Bogner: Data curation, formal analysis. K. Pedersen: Investigation. B. Tan: Investigation. N. Boice: Investigation. A. Acharya: Investigation. M. Abdiannia: Investigation. F. Gao: Formal analysis, visualization. H.H. Yoon: Investigation. M. Zhu: Investigation. N.A. Trikalinos: Investigation. L. Ratner: Investigation. O. Aranha: Investigation. W.G. Hawkins: Investigation. B.H. Herzog: Investigation. D.G. DeNardo: Investigation, writing-original draft, writing–review and editing.

NCI R01CA203890, Barnes Jewish Foundation, Precision Medicine Research Associates, Verastem, Merck. D.G. DeNardo and A. Wang-Gillam were supported by NCI R01CA177670, R01CA203890, P30CA09184215, and the BJC Cancer Frontier Fund. The trial was also supported by Verastem and Merck for the therapeutic agents defactinib and pembrolizumab, respectively.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

1.
Conroy
T
,
Desseigne
F
,
Ychou
M
,
Bouche
O
,
Guimbaud
R
,
Becouarn
Y
, et al
.
FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer
.
N Engl J Med
2011
;
364
:
1817
25
.
2.
Von Hoff
DD
,
Ervin
T
,
Arena
FP
,
Chiorean
EG
,
Infante
J
,
Moore
M
, et al
.
Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine
.
N Engl J Med
2013
;
369
:
1691
703
.
3.
Le
DT
,
Durham
JN
,
Smith
KN
,
Wang
H
,
Bartlett
BR
,
Aulakh
LK
, et al
.
Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade
.
Science
2017
;
357
:
409
13
.
4.
O'Reilly
EM
,
Oh
DY
,
Dhani
N
,
Renouf
DJ
,
Lee
MA
,
Sun
W
, et al
.
Durvalumab with or without tremelimumab for patients with metastatic pancreatic ductal adenocarcinoma: a phase 2 randomized clinical trial
.
JAMA Oncol
2019
;
5
:
1431
8
.
5.
Hu
ZI
,
Shia
J
,
Stadler
ZK
,
Varghese
AM
,
Capanu
M
,
Salo-Mullen
E
, et al
.
Evaluating mismatch repair deficiency in pancreatic adenocarcinoma: challenges and recommendations
.
Clin Cancer Res
2018
;
24
:
1326
36
.
6.
Hammel
P
,
Zhang
C
,
Matile
J
,
Colle
E
,
Hadj-Naceur
I
,
Gagaille
MP
, et al
.
PARP inhibition in treatment of pancreatic cancer
.
Expert Rev Anticancer Ther
2020
;
20
:
939
45
.
7.
Wong
W
,
Raufi
AG
,
Safyan
RA
,
Bates
SE
,
Manji
GA
.
BRCA mutations in pancreas cancer: spectrum, current management, challenges and future prospects
.
Cancer Manag Res
2020
;
12
:
2731
42
.
8.
Serrels
A
,
Lund
T
,
Serrels
B
,
Byron
A
,
McPherson
RC
,
von Kriegsheim
A
, et al
.
Nuclear FAK controls chemokine transcription, Tregs, and evasion of anti-tumor immunity
.
Cell
2015
;
163
:
160
73
.
9.
Frame
MC
,
Patel
H
,
Serrels
B
,
Lietha
D
,
Eck
MJ
.
The FERM domain: organizing the structure and function of FAK
.
Nat Rev Mol Cell Biol
2010
;
11
:
802
14
.
10.
Jiang
H
,
Hegde
S
,
Knolhoff
BL
,
Zhu
Y
,
Herndon
JM
,
Meyer
MA
, et al
.
Targeting focal adhesion kinase renders pancreatic cancers responsive to checkpoint immunotherapy
.
Nat Med
2016
;
22
:
851
60
.
11.
Stokes
JB
,
Adair
SJ
,
Slack-Davis
JK
,
Walters
DM
,
Tilghman
RW
,
Hershey
ED
, et al
.
Inhibition of focal adhesion kinase by PF-562,271 inhibits the growth and metastasis of pancreatic cancer concomitant with altering the tumor microenvironment
.
Mol Cancer Ther
2011
;
10
:
2135
45
.
12.
Infante
JR
,
Camidge
DR
,
Mileshkin
LR
,
Chen
EX
,
Hicks
RJ
,
Rischin
D
, et al
.
Safety, pharmacokinetic, and pharmacodynamic phase I dose-escalation trial of PF-00562271, an inhibitor of focal adhesion kinase, in advanced solid tumors
.
J Clin Oncol
2012
;
30
:
1527
33
.
13.
Jones
SF
,
Siu
LL
,
Bendell
JC
,
Cleary
JM
,
Razak
AR
,
Infante
JR
, et al
.
A phase I study of VS-6063, a second-generation focal adhesion kinase inhibitor, in patients with advanced solid tumors
.
Invest New Drugs
2015
;
33
:
1100
7
.
14.
Shimizu
T
,
Fukuoka
K
,
Takeda
M
,
Iwasa
T
,
Yoshida
T
,
Horobin
J
, et al
.
A first-in-Asian phase 1 study to evaluate safety, pharmacokinetics and clinical activity of VS-6063, a focal adhesion kinase (FAK) inhibitor in Japanese patients with advanced solid tumors
.
Cancer Chemother Pharmacol
2016
;
77
:
997
1003
.
15.
Gilabert
M
,
Chanez
B
,
Rho
YS
,
Giovanini
M
,
Turrini
O
,
Batist
G
, et al
.
Evaluation of gemcitabine efficacy after the FOLFIRINOX regimen in patients with advanced pancreatic adenocarcinoma
.
Medicine (Baltimore)
2017
;
96
:
e6544
.
16.
Viaud
J
,
Brac
C
,
Artru
P
,
Le Pabic
E
,
Leconte
B
,
Bodere
A
, et al
.
Gemcitabine as second-line chemotherapy after Folfirinox failure in advanced pancreatic adenocarcinoma: A retrospective study
.
Dig Liver Dis
2017
;
49
:
692
6
.
17.
Portal
A
,
Pernot
S
,
Tougeron
D
,
Arbaud
C
,
Bidault
AT
,
de la Fouchardiere
C
, et al
.
Nab-paclitaxel plus gemcitabine for metastatic pancreatic adenocarcinoma after Folfirinox failure: an AGEO prospective multicentre cohort
.
Br J Cancer
2015
;
113
:
989
95
.
18.
Vogel
A
,
Rommler-Zehrer
J
,
Li
JS
,
McGovern
D
,
Romano
A
,
Stahl
M
.
Efficacy and safety profile of nab-paclitaxel plus gemcitabine in patients with metastatic pancreatic cancer treated to disease progression: a subanalysis from a phase 3 trial (MPACT)
.
BMC Cancer
2016
;
16
:
817
.
19.
Tang
KJ
,
Constanzo
JD
,
Venkateswaran
N
,
Melegari
M
,
Ilcheva
M
,
Morales
JC
, et al
.
Focal adhesion kinase regulates the DNA damage response and its inhibition radiosensitizes mutant KRAS lung cancer
.
Clin Cancer Res
2016
;
22
:
5851
63
.
20.
Jiang
H
,
Liu
X
,
Knolhoff
BL
,
Hegde
S
,
Lee
KB
,
Jiang
H
, et al
.
Development of resistance to FAK inhibition in pancreatic cancer is linked to stromal depletion
.
Gut
2020
;
69
:
122
32
.
21.
Shapiro
IM
,
Kolev
VN
,
Vidal
CM
,
Kadariya
Y
,
Ring
JE
,
Wright
Q
, et al
.
Merlin deficiency predicts FAK inhibitor sensitivity: a synthetic lethal relationship
.
Sci Transl Med
2014
;
6
:
237ra68
.