br Trial Design and Statistical Analysis
Trial Design and Statistical Analysis
The primary objectives of the protocol were to determine if it is feasible to safely administer 6 doses of AMP-224 in combination with Charalampos S. Floudas et al
radiation to patients with mCRC and to explore effects of treatment on immunologic parameters. All patients enrolled in the trial were considered as 1 group. For the feasibility outcome, a goal of 11 of 15 evaluable patients able to safely receive 6 doses of AMP-224 was determined to be consistent with an 80% goal of administration and consequently considered a successful outcome for the trial. We esti-mated 95% confidence intervals (CIs) for response rates using the Wilson method.38 OS and PFS beginning at Malonyl Coenzyme A 1, day 1 of treatment were estimated using the KaplaneMeier method,39 and presented along with medians and 95% CIs for PFS and OS. Data were analyzed with R40 using the Rstudio environment41; the Hmisc package42 was used for CIs. SAS version 9.4 (SAS Institute, Cary, NC) was used to perform the survival analyses.
Between November 2014 and August 2016, 17 patients were assessed for eligibility and 15 patients were included. Six patients were assigned to the DL1 arm and 9 patients to the DL2 arm (Figure 1). All 15 patients were treated with at least 1 dose of AMP-224 and were included in the analysis. The baseline characteristics of the participants are listed in Table 1. Of the 15 patients, 93% had metastatic cancer at diagnosis, and 80% had undergone a metastasectomy. The median number of previous chemotherapy lines was 3 (range, 2-5). Four (27%) patients had microsatellite stable (MSS) tumors tested using immunohistochemistry, which was not available for the remaining 11 patients. KRAS status was available in 12 (80%) of patients, of whom 8 (53%) had KRAS-mutated tumors.
Feasibility and Safety
As of data cutoff on March 7, 2018, no participant was still receiving the treatment. Ten (67%) patients received all 6 doses of AMP-224, and 1 patient (7%) continued treatment with AMP-224 for an additional 4 cycles before disease progression. Treatment was discontinued before completion of the planned 6 doses of AMP-224 in 5 (33%) patients because of progressive disease. There was no treatment discontinuation because of AEs. No DLT was encountered.
The safety population consisted of 15 patients. All patients experienced at least 1 AE, with 9 (60%) experiencing treatment-related AEs, all of which were of Grade 1 or 2. Treatment-related toxicities are summarized in Table 2, which excludes toxicities directly attributable to the SBRT or the biopsy (eg, pain) according to standard of care experience. Infusion reactions were the most common treatment-related AEs, occurring in 6 (40%) patients, which were managed accordingly.
Clinical Colorectal Cancer Month 2019 - 3
Anti-PD-1 and Radiation in Colorectal Cancer
Figure 1 Consolidated Standards of Reporting Trials Diagram
Enumeration of Immune Cells
Ten paired pre- and post-treatment tumor biopsies were available from 5 patients. After quality control analysis, 1 sample had a low count of mapped reads and was excluded with its pair, leaving 8 paired samples for further analysis, derived from 1 patient with stable disease as the best response and 3 patients with progressive disease as the best response. Differential gene expression analysis results are presented in Supplemental Figure 1 in the online version. Immune infiltrate CIBERSORT (https://cibersort.stanford.edu) analysis showed P values < .05 for 3 of 4 pretreatment samples; in 2 of these 3 samples, there was a predominance of M2 macrophages and comparatively little T-cell response (see Supplemental Figure 2 in the online version). All post-treatment samples had P values >
.05, which are summarized in Figure 3 and are reported descriptively and considered exploratory.