Hurwitz, Herbert I.

TPS873 Background: Initial treatment of metastatic colorectal cancer (mCRC) involves a 5-fluorouracil based chemotherapy regimen, often in combination with anti-VEGF therapy. Upon disease progression, multiple studies have suggested a benefit for continued anti-VEGF therapy. There is increasing evidence that VEGF plays a role in cancer immune evasion. Targeting of inflammatory and immune checkpoints are attractive approaches to enhance the benefits of anti-VEGF therapy. The safety and activity of the anti-PD-L1 antibody atezolizumab with bevacizumab and with 5-FU, oxaliplatin, and bevacizumab have been recently reported; the combinations were well-tolerated and suggested clinical activity. Therefore, bevacizumab may increase the immunogenicity of colorectal cancers, and the combination of atezolizumab plus bevacizumab may be associated with clinically meaningful response rates and disease control. Furthermore, there may be candidate biomarkers that may identify those patients most likely to benefit from this combination. Methods: In this randomized, double-blind, multicenter, placebo-controlled phase II study, 135 patients with mCRC will be randomized 2:1 to receive capecitabine/bevacizumab/atezolizumab or capecitabine/bevacizumab/placebo. Patients with prior PD-L1/PD-1 therapy are ineligible. Primary and secondary efficacy will be conducted using ITT analysis. Safety analyses will include all randomized patients who receive at least one dose of study treatment. The primary objective is PFS. The secondary objectives are ORR, OS, safety, and tolerability. The primary comparison will be superiority of the active treatment for the PFS endpoint, atezolizumab versus placebo testing at 1-sided α = 0.1 (log rank test). A PFS hazard ratio of 0.618 (active treatment versus placebo) is detectable with 86% power (1 sided α = 0.1). No interim analyses for futility or efficacy will be conducted. A subset analysis will be performed in the microsatellite-high patients looking at potential differences in PFS and OS. Trial is active and currently recruiting patients. Clinical trial information: NCT02873195.

726 Background: Primary tumor location is an important prognostic factor in mCRC. Efficacy differences with EGFR antibodies by tumor location show a predictive effect (Arnold et al. Ann Oncol 2017). Two first-line (1L) pivotal bev trials showed the prognostic value of sidedness (Loupakis et al. J Natl Cancer Inst 2015). In this exploratory study, further analysis of the predictive effect of bev is presented. Methods: Data from two randomized phase III studies (NO16966 [NCT00069095], AVF2107g [NCT00109070]) of bev+CT vs CT-alone as 1L treatment for mCRC were analyzed retrospectively. Patients (pts) with sidedness information available were included; pts from NO16966 (part 1) randomized to CT only were excluded. Sidedness was identified by case report forms. Results: Sidedness was determined in 1590 pts (27% Right and 73% Left) out of 2214 pts total (Table). In the pooled analysis, PFS improved with bev+CT compared to CT only in both right-sided (HR=0.75; 95% CI 0.61, 0.93) and left-sided (HR=0.76; 95% CI 0.67, 0.86) mCRC with similar magnitude of benefit in both sides. OS also improved with bev+CT in right-sided (HR=0.82; 95% CI 0.65, 1.03) and left-sided (HR=0.85; 95% CI 0.74, 0.98) pts. Conclusions: This exploratory subgroup analysis of two pivotal studies indicates that the effect of bev is independent of tumor location in mCRC patients. Survival outcomes. Clinical trial information: NCT00069095 and NCT00109070. [Table: see text]

617 Background: There is limited, often conflicting evidence about AE timing, severity or associations with outcomes with the use of cytotoxic agents in cancer treatment. We investigated the impact on overall survival (OS) and progression-free survival (PFS) of selected common AEs (neutropenia, diarrhea, nausea, vomiting, neuropathy) occurring in patients receiving first line oxaliplatin (Oxa)- and/or irinotecan(Iri)-based regimens for mCRC. Methods: The CTCAE grading scores of at least one AE of interest were available on 9812 pts treated with chemotherapy alone (median age 63; 62.4% male, 50.1% ECOG PS 0) from 17 1st-line randomized trials. Patients who also received biologics were excluded in the primary analyses. AEs occurring during the first 6 weeks of treatment and entire treatment were analyzed by stratified multivariable Cox models in relationship to OS/PFS. 55.7% pts received Oxa- regimens, 35.7% Iri-regimens, and 8.6% combined Oxa- and Iri-regimens. Results: Within the first 6 weeks of treatment, G3+ neutropenia (HRadj= 1.3, 95% CI, 1.06-1.59, padj 0.01), diarrhea (HRadj= 1.48, 95% CI, 1.23-1.79, padj < .0001), nausea (HRadj= 1.53, 95% CI, 1.17-1.99, padj 0.002) and vomiting (HRadj= 1.56, 95% CI, 1.18-2.07, padj 0.002) were associated with significantly worse OS for Iri-regimens, but only G3+ nausea predicted for worse OS for Oxa- regimens (HRadj= 1.61, 95% CI, 1.18-2.21, padj 0.003). For AEs experienced at any time, G3+ neutropenia and neuropathy were significantly associated with longer PFS and OS for Oxa-regimens, while G3+ vomiting and nausea were associated with worse OS for both Oxa- and Iri-based regimens. Sensitivity analysis showed largely concordant results by including pts who also received biologics. Conclusions: The association between more severe selected AEs and outcome varies between AEs and is influenced by timing of the occurrence. More severe selected AEs occurring early in treatment are associated with worse outcomes. In contrast, for AEs occurring at any time, G3+ neutropenia and neuropathy predicted for longer PFS and/or OS in Oxa-treated pts.

612 Background: mCRC is a heterogeneous disease leading to possible disparate responses among lesions in a single pt. This study assesses the heterogeneity of lesion responses and evaluates the prognostic value of LBR12 and its mortality risk discrimination beyond RECIST responses. Methods: Pts with ≥ 2 lesions and no new lesions at 12 weeks (wks) were eligible. For each pt, after 12 wks of Rx, each lesion was categorized as progressive disease (L-PD: ≥20% increase in the longest diameter [LD]), partial response (L-PR: ≥30% decrease in LD), or stable disease (L-SD: neither L-PD nor L-PR). LBR12 was defined, per patient, by the combination of lesion responses: homogeneous LBR12 (L-PD, L-SD, L-PR only) and heterogeneous LBR12 (mixture of L-PD/L-SD, L-PD/L-PR, L-SD/L-PR). LBR12 and overall survival (OS) were correlated using stratified multivariate Cox models after adjusting for age, gender, and ECOG PS. Results: Among 9,092 mCRCs (Rx: chemo alone 44%; chemo + VEGF inhibitor [VEGFi] 42%; chemo + EGFR inhibitor [EGFRi] 10%) from 16 1st-line studies. Median OS: 2.2 years. Per RECIST at 12 wks, CR 0%; PR 36.1%; SD 60.9%; PD 3.0%. Responses in 52% were heterogeneous (Table). VEGFi and EGFRi treated pts had the highest rate of L-SD/L-PR (45%) and L-PR only (22%) status, respectively. Median OS increased monotonically across pts with more L-PRs and fewer L-PDs (Table). Pts with L-SD/L-PR status, among which 51% had SD per RECIST, had longer OS than those with L-SD only status (HRadj.= .81, padj.< .0001), but shorter OS than those with L-PR only status (HRadj.= 1.46, padj.< .0001). Pts with L-PD/L-SD status, among which 71% were SD per RECIST, had shorter OS than those with L-SD only status (HRadj.= 2.22, padj.< .0001). These associations were consistent across treatment regimens. Conclusions: The lesion-based response captures the heterogeneity of within pt lesion responses and provides refinement in predicting outcome beyond RECIST response at 12 wks. [Table: see text]

11546 Background: TRC105, an endoglin-targeting monoclonal antibody with anti-angiogenic and anti-tumor activity, is being evaluated in multiple diseases. Here we report on pharmacodynamic and prognostic biomarkers in patients (pts) treated with both TRC105 and anti-VEGF agents. Methods: Plasma samples were collected from pts on three phase 2 trials combining TRC105 with an anti-VEGF agent: axitinib in metastatic renal cell carcinoma (mRCC), pazopanib in advanced soft tissue sarcoma, and bevacizumab in glioblastoma (GBM). Baseline and on-treatment levels of 22 soluble protein biomarkers were assessed. Results: Soluble endoglin markedly increased after TRC105 treatment in all pts ( p< 0.001) as previously reported. BMP9 (a ligand for endoglin) and TGFβR3 (a type III TGFβ receptor) decreased in sarcoma pts at Cycle 2 Day 1 (C2D1) and generally remained below baseline throughout the course of treatment (BMP9, p= 0.004; TGFβR3, p= 0.003). Although TGFβR3 was decreased at C2D1 in mRCC (p = 0.030), no clear patterns were observed over time. Overall BMP9 levels did not change in response to therapy in either mRCC or GBM. Osteopontin (OPN) levels, a downstream effector of TGFβ signaling, were increased in sarcoma pts [ p= 0.002 at C2D15, p< 0.001 at C4D1 and end of study (EOS)]; however, in mRCC ( p= 0.010) and GBM ( p= 0.003), OPN was only elevated at EOS. Increases in PlGF and VEGFD, and decreases in VEGFR2 were observed across all studies, as previously noted for VEGF inhibitors. In the mRCC trial, 5 of 18 pts exhibited a ≥30% tumor reduction. Lower OPN ( p= 0.026) and higher TGFβR3 ( p= 0.003) levels at baseline correlated with radiographic response to treatment. In the sarcoma trial, 6 of 19 pts responded (CHOI criteria) in which lower baseline levels of ICAM1 ( p= 0.018) and TSP2 ( p= 0.042) correlated with stable disease. Conclusions: In these trials, increases of soluble endoglin in response to TRC105 were observed, independent of the presence of any specific VEGF inhibitor. Differential regulation of BMP9, TGFβR3, and OPN suggests potential disease-specific modulation of key TGFβ signaling molecules in response to dual therapy. Baseline levels of OPN and TGFβR3 showed potential prognostic value in mRCC. Confirmation in larger trials is needed.

3582 Background: The prognostic importance of the incidence, severity, type and duration of AEs pts experience during chemotherapy varies between tumor types, and the available evidence across the board is often conflicting. Here we investigated the impact of the overall severity of AEs among pts with mCRC receiving first-line oxaliplatin (Oxa)- and/or irinotecan(Iri)-based regimens. Methods: The overall severity of AE data (i.e., max grade (G) of all AEs) were available on 3,971 pts (median age 61; 60% male, 47% ECOG PS 1+; 57% 2+ metastatic sites) enrolled onto 6 1st-line randomized trials. Around 46%, 45%, and 9% of pts had received Oxa-, Iri-, and Oxa+Iri-based regimens, respectively. Pts receiving biologic agents were excluded. Stratified multivariate Cox models were used to assess the associations with overall survival (OS) and progression-free survival (PFS); adjusted hazard ratios (HRadj) and 95% confidence intervals (CIs) are reported. Results: Pts who only received Oxa-based treatment reported the lowest rate of G3+ AEs (p < .0001) compared to pts treated with Iri- or Oxa+Iri-based regimens. Older age, female gender, and PS 1 or 2+ were associated with higher grade AEs (all p < .0001). Considering AEs experienced within 6w after randomization, 10% and 61% of pts experienced G4+ and G2-3 AEs, respectively. G3+ AEs were associated with a shorter OS for both pts receiving Oxa- (HRadj= 1.2, 95% CI, 1.1-1.3, padj < .0001) and Iri-based regimens (HRadj= 1.4, 95% CI, 1.2-1.5, padj < .0001). For the entire treatment course, 19% and 72% of pts experienced G4+ and G2-3 AEs, respectively. For Oxa-based regimens, pts with G3+ AEs had a longer OS (HRadj= 0.86, 95% CI, 0.78-0.94, padj = .0016), whereas G3+ AEs were associated with a shorter OS (HRadj= 1.2, 95% CI, 1.1-1.4, padj = .0004) for pts treated with Iri-based regimens. Similar patterns were seen for PFS. Conclusions: Pts who reported higher grade AEs during initial treatment (≤6w) have significantly worse outcome than those who do not. Further analyses with treatment exposure/detailed dose-AE profile and its impact on survival are warranted.

3535 Background: CRC is known to be a heterogeneous disease. This study quantifies within pt heterogeneity in early lesion change rate (LCR) in the era of targeted agents compared to chemo alone and its potential impact on survival. Methods: Pts with 2-10 lesions measured at baseline were included. For each lesion, the early LCR was defined as the change in size (shrinking or growing) from baseline to 12 weeks on treatment. Within pt heterogeneity in early LCR among lesions was estimated by standard deviation (STD). A larger value of STD indicates larger variation of LCR per pt. Stratified multivariate Cox models were used to assess the associations between LCR STD with overall survival (OS). Adjusted hazard ratios (HRadj) and 95% confidence intervals (CIs) are reported. Results: Data were available on 9,092 mCRC pts (median age 61; 60% male, 55% ECOG PS 0; 61% 2+ metastatic sites) enrolled in 16 1st-line randomized trials, with 44%, 42%, and 10% of pts received chemo alone, + a VEGF inhibitor (VEGFi) or an EGFR inhibitor (EGFRi), respectively. LCR heterogeneity is the highest among pts received EGFRi but lowest among pts received VEGFi (Table). Overall, higher heterogeneity is associated with worst OS (HRadj1.22, 95% CI (1.16, 1.27)). The effect is most pronounced in pts received VEGFi (interaction p=0.0012). Conclusions: There was heterogeneity observed in lesion size changes within pts. Its magnitude varies across treatment approaches, and was associated with poor survival. This preliminary result reveals the great potentials to define novel response endpoint and refine treatment decision-making by incorporating heterogeneities in lesion changes. [Table: see text]