A phase II trial of cabozantinib and erlotinib for patients with EGFR and c-Met co-expressing metastatic pancreatic adenocarcinoma (PDAC).

e16764 Background: Both EGFR and the c-MET receptors are overexpressed in a majority of PDACs. Inhibition of both receptors simultaneously may be required for anti-tumor activity. Erlotinib, an EGFR inhibitor, has modest activity in metastatic PDAC and is approved by the FDA in combination with gemcitabine. Cabozantinib is a tyrosine kinase inhibitor targeting AXL, FLT-3, KIT, MER, MET, RET, ROS1, TIE-2, TRKB, TYRO3, and VEGFR-1, -2, and -3. Preclinical data suggests that the addition of cabozantinib to erlotinib leads to tumor shrinkage and improvement in survival in a KPC PDAC mouse model compared to gemcitabine alone. This phase II study tests this hypothesis in patients with metastatic PDAC that co-express c-MET and EGFR. Methods: Key eligibility includes patients (pts) with metastatic PDAC with EGFR and c-MET overexpression (as determined by centrally tested IHC of 2+ or greater) that have progression on one prior chemotherapy regimen. Patients were treated with cabozantinib (40mg daily) and erlotinib (100mg daily) continuously. This dosing is based on previous combination data in NSCLC. This is a single arm two-stage phase II study with a primary endpoint of overall response rate. Secondary endpoints include of PFS, DCR and OS. Results: From October 2017 to October 2019, 43 pts were screened with 7 pts (median age 62 [range 51-76)] enrolled and treated on study. Pts had a median of 1 line of prior systemic chemotherapy. Most common reason for screen failure was due to lack of co-expression of c-MET and EGFR. EGFR IHC expression was +2 in 4 pts, +3 in 3 pts; c-MET IHC expression was +2 in 5 pts and +3 in 2 pts. Most common any-grade adverse events attributable to cabozantinib and erlotinib include: diarrhea (71%), AST increase (43%), fatigue (43%), nausea (43%), and rash (43%). Only one grade 3 event of fatigue occurred. All pts had clinical and/or radiographic progression within 1-2 months after initiating study therapy. Conclusions: The combination of cabozantinib and erlotinib was well tolerated with manageable toxicity. Due to lack of clinical responses, this study has been terminated due to futility. Clinical trial information: NCT03213626 .

Predictive impact of PD-L1-expressing circulating tumor cells in NSCLC patients treated with nivolumab.

11541 Background: PD-L1 expression on tumor tissue is associated with response to PD-1 blockade in NSCLC. Here, we conducted a serial evaluation of PD-1-expressing circulating tumor cells (CTCs) as a potential real-time diagnostic modality in NSCLC patients treated with nivolumab. Methods: Advanced NSCLC patients after failure of at least one prior chemotherapy regimen received nivolumab monotherapy (3mg/kg, q2W) until progressive disease (PD) or unacceptable toxicity. Peripheral whole blood (3 mL) was collected for CTC evaluation at baseline and at week 4. CTCs were detected using microcavity array system (Hitachi Chemical Co., Ltd, Chikusei, Japan). PD-L1 expression was immunohistochemically examined on both tumor tissues and CTCs. This study was registered at UMIN (ID: 000024414). Results: Thirty patients were registered in the study between January 2016 and September 2016 at Wakayama Medical University Hospital and 29 were included in the analysis. Demographics of the patients were as follows: median age 70 (range, 49 to 86); male 73 %; stage IV, 100 %; squamous/non-squamous, 27/73 %. At baseline, CTCs were detected in all patients (median, 15; range, 1 to 90) and PD-L1-expressing CTCs were detected in 87% of patients. Tumor proportion score (TPS) of PD-L1 expression on CTCs ranged from 6% to 100%, indicating intrapatient heterogeneity. Matched tumor tissues were available from 14 patients and 7 showed the PD-L1 TPS ≥ 50%. No positive correlation was observed on PD-L1 expression between tumor tissues and CTCs based on TPS (R2 = 0.0035). Overall response rate was 25% (7/29), and disease control rate was 54% (15/29). Total CTC count was significantly decreased after nivolumab treatment at week 4 (p < 0.05), but no significant change was observed in PD-L1 TPS on CTC. Patients harboring CTCs with PD-L1 TPS 50% or more at baseline were significantly more likely to achieve non-PD than those harboring CTCs with TPS less than 50% (p < 0.05). Conclusions: This is the first report on a serial monitoring of PD-L1 expression on CTCs in patients treated with nivolumab. PD-L1-expressing CTCs are suggested to hold potential for predicting clinical benefit. Clinical trial information: 000024414.

Epacadostat plus pembrolizumab in patients with SCCHN: Preliminary phase I/II results from ECHO-202/KEYNOTE-037.

6010 Background: Indoleamine 2,3-dioxygenase 1 (IDO1) is a tryptophan-catabolizing enzyme that induces immune tolerance by T-cell suppression. IDO1 overexpression has been associated with poor survival in SCCHN. Epacadostat (E) is a potent, selective oral IDO1 inhibitor. ECHO-202/KEYNOTE-037 is an open-label, phase I/II (P1/2) study evaluating E plus PD-1 inhibitor pembrolizumab (P) in multiple tumor types. We report preliminary P1/2 efficacy, safety, and tolerability findings in the SCCHN cohort as of a 29OCT2016 data cutoff. Methods: Eligible adult patients (pts) had metastatic SCCHN and received ≥1 prior chemotherapy regimen that included a platinum agent. Prior checkpoint inhibitor therapy (tx) was not permitted, and pts with carcinoma of the nasopharynx or salivary gland were excluded. In P1 dose escalation (3+3+3), pts received E (25, 50, 100, or 300 mg PO BID) + P (2 mg/kg or 200 mg IV Q3W); MTD was not exceeded. E (100 mg BID) + P (200 mg Q3W) dosing was selected for P2 cohort expansion. Response was assessed in RECIST 1.1 evaluable pts. Results: A total of38 pts (P1, n = 2; P2, n = 36) were evaluated. Median age was 63 years, 87% of pts were men, 95% were white, and 66% received prior cetuximab. Of 36 efficacy-evaluable pts, 81% (n = 29) received 1–2 prior lines of tx and 19% (n = 7) received ≥3 prior lines of tx. ORR (CR+PR) and DCR (CR+PR+SD) for pts with 1–2 prior tx were 34% (2 CR, 8 PR) and 62% (8 SD), respectively; for pts with ≥3 prior tx, ORR and DCR were 14% (1 PR) and 43% (2 SD). Response was observed regardless of HPV status. At data cutoff, 9/11 responses were ongoing (range, 1+ to 563+ days). PFS and biomarker analyses are ongoing. The most common TRAEs in all 38 pts were fatigue (24%), nausea (11%), and decreased weight (11%). Grade ≥3 TRAEs occurred in 11% of pts; only increased amylase and lipase (both asymptomatic) were grade ≥3 TRAEs that occurred in > 1 pt. TRAEs led to discontinuation in 1 pt (increased amylase and lipase). Conclusions: In pts with advanced SCCHN, E + P was generally well tolerated and associated with encouraging response rates, particularly in pts with 1–2 prior lines of tx. A phase III SCCHN study is planned. Clinical trial information: NCT02178722.

Two randomized, placebo-controlled phase 3 studies of ruxolitinib (Rux) + capecitabine (C) in patients (pts) with advanced/metastatic pancreatic cancer (mPC) after failure/intolerance of first-line chemotherapy: JANUS 1 (J1) and JANUS 2 (J2).

343 Background: The JAK-STAT pathway plays a role in the pathogenesis of PC. In a phase 2 study, the JAK1/2 inhibitor Rux + C improved OS vs C alone in mPC pts with evidence of systemic inflammation. Methods: In two phase 3 studies, advanced/mPC pts with only 1 prior chemotherapy regimen and C-reactive protein > 10mg/L were randomized to 21-day cycles of Rux 15 mg BID + C 2000 mg/m2/d (day 1-14) vs placebo (Pb) BID + C 2000 mg/m2/d (day 1-14). The primary endpoint was OS. The studies were terminated early based on efficacy findings of a planned interim analysis for J1, with J2 enrollment incomplete. Results: 321 pts were randomized in J1 and 86 in J2. The median age was similar between groups (Rux + C vs Pb + C): J1, 68 vs 67 yrs; J2, 68 vs 69 yrs. The most common prior chemotherapy regimens were gemcitabine (gem) + nab-paclitaxel (J1: 50% vs 51%, J2: 42% vs 37%), gem alone (J1: 24% vs 21%, J2: 28% vs 33%). Median study treatment durations were 42 d for both Rux and Pb in J1; 40 d for Rux and 43 d for Pb in J2. Discontinuations were mainly due to disease progression (J1: 62% vs 65%, J2: 51% vs 58%), AEs (J1: 4% vs 10%, J2: 12% vs 12%) or death (J1: 8% vs 5%, J2: 7% vs 5%). Rux + C did not show a benefit over C alone in OS or PFS (see Table). Overall response rates were < 5% in each treatment group (NS).The most common grade 3/4 non-hematologic AEs were abdominal pain (J1: 7% vs 14%; J2: 14% vs 5%), palmar-plantar erythrodysesthesia syndrome (J1: 7% vs 4%; J2: 0% vs 12%), pulmonary embolism (J1: 4% vs 3%; J2: 10% vs 5%), dyspnea (J1: 2% vs 2%; J2: 10% vs 2%) and vomiting (J1: 5% vs 5%; J2: 2% vs 12%). The most common new/worsening hematologic labs abnormalities of grade 3/4 included: anemia (J1: 20% vs 6%; J2: 11% vs 8%) and lymphopenia (J1: 17% vs 9%; J2: 16% vs 8%). Conclusions: When added to C, Rux was well tolerated but did not improve clinical outcomes in the second-line treatment of mPC pts. Clinical trial information: NCT02117479 and NCT02119663. [Table: see text]

Survival and long-term follow-up of the phase I trial of nivolumab (Anti-PD-1; BMS-936558; ONO-4538) in patients (pts) with previously treated advanced non-small cell lung cancer (NSCLC).

8030 Background: The immune checkpoint receptor programmed death-1 (PD-1) negatively regulates T-cell activation. Nivolumab, a PD-1 receptor blocking antibody, was evaluated in a phase 1 study in pts with various tumors including NSCLC (Topalian et al, NEJM 2012;366:2443). Methods: Pts with ≥1 prior chemotherapy regimen received nivolumab (1-10 mg/kg IV Q2W) for ≤12 cycles (4 doses/8W cycle) or until discontinuation criteria were met. We report initial overall survival (OS) and updated safety and response data for NSCLC pts. Results: 127 pts evaluable for safety received nivolumab at 1, 3, or 10 mg/kg as of July 2012. Common drug-related AEs were decreased appetite (9%), anemia (8%), diarrhea, nausea, and pruritus (7% each). The most common G3/4 AEs were fatigue, pneumonitis, and elevated AST (2% each). Two drug-related deaths from pneumonitis occurred early in the trial and led to increased monitoring without further deaths from pneumonitis. Median OS (mOS) across all dose cohorts was 9.2 mo and 9.6 mo for squamous (sq) and non-sq NSCLC, respectively. mOS was not reached at the 3 mg/kg dose (phase 3 dose) for either histology. Sustained OS was observed, with 44%/ 41% and 44%/ 17% of pts (sq/non-sq) alive at 1 and 2 years, respectively (Table). Prolonged ORs occurred in both histologies (Table). Conclusions: In this long-term follow-up of a phase I trial, nivolumab had an acceptable safety profile and showed an encouraging sustained OS benefit across histologies in previously treated advanced NSCLC. Follow-up through a Feb 2013 data cut (≥1 yr follow-up for all pts) will be provided at presentation. Clinical trial information: NCT00730639. [Table: see text]

Nintedanib (BIBF 1120) plus docetaxel in NSCLC patients progressing after one prior chemotherapy regimen: Results of Lume-Lung 1, a randomized, double-blind, phase III trial.

LBA8011 The full, final text of this abstract will be available at abstract.asco.org at 7:30 AM (EDT) on Monday, June 3, 2013, and in the Annual Meeting Proceedings online supplement to the June 20, 2013, issue of Journal of Clinical Oncology. Onsite at the Meeting, this abstract will be printed in the Monday edition of ASCO Daily News.

Multinational, Double-Blind, Phase III Study of Prednisone and Either Satraplatin or Placebo in Patients With Castrate-Refractory Prostate Cancer Progressing After Prior Chemotherapy: The SPARC Trial

Purpose This multinational, double-blind, randomized, placebo-controlled, phase III trial assessed the efficacy and tolerability of the oral platinum analog satraplatin in patients with metastatic castrate-refractory prostate cancer (CRPC) experiencing progression after one prior chemotherapy regimen. Patients and Methods Nine hundred fifty patients were randomly assigned (2:1) to receive oral satraplatin (n = 635) 80 mg/m2 on days 1 to 5 of a 35-day cycle and prednisone 5 mg twice daily or placebo (n = 315) and prednisone 5 mg twice daily. Primary end points were progression-free survival and overall survival (OS). The secondary end point was time to pain progression (TPP). Results A 33% reduction (hazard ratio [HR] = 0.67; 95% CI, 0.57 to 0.77; P < .001) was observed in the risk of progression or death with satraplatin versus placebo. This effect was maintained irrespective of prior docetaxel treatment. No difference in OS was seen between the satraplatin and placebo arms (HR = 0.98; 95% CI, 0.84 to 1.15; P = .80). Compared with placebo, satraplatin significantly reduced TPP (HR = 0.64; 95% CI, 0.51 to 0.79; P < .001). Satraplatin was generally well tolerated, although myelosuppression and GI disorders occurred more frequently with satraplatin. Conclusion Oral satraplatin delayed progression of disease and pain in patients with metastatic CRPC experiencing progression after initial chemotherapy but did not provide a significant OS benefit. Satraplatin was generally well tolerated. These results suggest activity for satraplatin in patients with CRPC who experience progression after initial chemotherapy.

Efficacy and safety of PF-00299804 (PF299) in patients (pt) with advanced NSCLC after failure of at least one prior chemotherapy regimen and prior treatment with erlotinib (E): A two-arm, phase II trial

8063 Background: Options are limited for pts with NSCLC following chemotherapy and E. PF299 is an oral irreversible small molecule inhibitor of the HER-1,-2, and -4 tyrosine kinases. Based on non-clinical and phase I NSCLC data, this ongoing phase II U.S. trial evaluates PF299 in pts with NSCLC (KRAS wild-type) who have progressive disease after at least 1 prior chemotherapy regimen and after E. Methods: Pts were enrolled by histology: adenocarcinoma (Arm A) and non-adenocarcinoma (Arm B), and received PF299 45 mg QD. Endpoints include objective response rate, duration of response, progression-free survival, survival, safety/tolerability, and pharmacokinetics. Pharmacodynamic endpoints include assessment of serum levels of HER2 and EGFR extracellular domains. Tissue and blood KRAS assays, and EGFR studies on available tissue, are also being performed. Forty-four and 22 response-evaluable pts were planned to be enrolled into Arms A and B respectively. Results: Thirty-four pts with progressive NSCLC (25 female, 14 smoker) have enrolled to date (Arm A: 30; Arm B: 4): median duration of prior E: 11.5 months; median time since E: 2.5 months. Among 20 response-evaluable pts, stable disease (SD) was observed in 9/18 pts in Arm A, and 1/2 pts in Arm B: median duration of SD: 11.5 weeks [range 6+, 32+ weeks]. Observation of disease control included pts who had recently (≤8 weeks) discontinued E; and also pts whose tumor had known EGFR T790M mutations. At time of data cutoff, confirmation per RECIST of 2 partial responses is pending. The most common treatment-related AEs were skin and gastrointestinal disorders, with grade 3 AEs in 19% and 13% of pts, respectively. Two pts experienced grade 4 pulmonary embolus/dyspnea deemed possibly treatment-related, both in the setting of progressive disease. Conclusions: PF299 shows encouraging activity in NSCLC pts after failure of prior chemotherapy and E. The AE profile was predictable and consistent with the prior phase I trial. At submission, enrollment in the adenocarcinoma arm is complete and enrollment in the non-adenocarcinoma arm continues; updated efficacy data and tumor characterization will be presented. [Table: see text]

Risk factors for severe cancer chemotherapy toxicity in the outpatient setting: Results of a nested case control study using a novel patient cohort

6545 Background: Risk factors for severe cancer chemotherapy-related toxicity (resulting in hospitalization and/or death) in the adult outpatient setting are largely unknown. Between 2003–2006, we identified all chemotherapy patients from a single hospital-affiliated outpatient cancer center who required admission to the hospital, and using a structured process that included monthly peer review meetings, we determined which admissions were due to chemotherapy toxicity. Among 2,082 patients who received chemotherapy, 185 required at least one admission. In this study we describe the development of a prediction model for severe chemotherapy-related toxicity. Methods: We undertook a nested case control study by randomly selecting 98 patients admitted to the hospital within 30 days of their last chemotherapy administration and 173 controls. We selected 33 clinical and demographic variables that we hypothesized might predict severe toxicity, and used standard descriptive, univariate and regression methods to characterize the data and develop a prediction model. Results: We were able to identify the following variables that predicted for severe chemotherapy toxicity: increasing age; increasing Charlson Comorbidity index (CCI); treatment with a prior chemotherapy regimen; prior admission for chemotherapy toxicity; decreasing hemoglobin and absolute neutrophil count (ANC) and increasing creatinine, and albumin. Independent variables that did not predict for increased risk for severe toxicity included the following: performance status; chemotherapy intent (curative, adjuvant or palliative); full dose versus reduced dose for cycle I; cisplatin-based chemotherapy; median income (by postal codes). Using logistic regression we created a parsimonious model (ROC AUC of 0.69) that included four clinical independent variables: CCI; prior chemotherapy regimen; ANC; and albumin. Conclusions: This analysis may permit the development of a predictive model which will include CCI, prior chemotherapy regimen, ANC, and albumin to help identify patients prior to the initiation of chemotherapy who are risk for severe toxicity. No significant financial relationships to disclose.