br Of the genes differentially expressed see Supplemental Ta
Of the 1043 AMG 925 differentially expressed (see Supplemental Table 1 in the online version), genes with P values < .05 that were submitted to the Webserver, 893 genes were included in the GSEA. Of the 50 top overlapping gene sets (see Supplemental Table 2 in the online version), there was a predominance of the immune signature gene sets, with almost all of the 50 overlapping gene sets from that collection.
In this single-center pilot study, we explored the feasibility, safety, and efficacy of AMP-224 treatment in combination with low-dose cyclophosphamide and SBRT in patients with mCRC in whom standard treatments had failed.
In our study, the treatment was well tolerated. All patients expe-rienced at least 1 AE, and 9 (60%) experienced treatment-related AEs, all of which were Grade 1 or 2. No toxicity-related deaths and Grade 3 or 4 AEs were observed. The overall toxicity profile for AMP-224 used in combination with low-dose cyclophosphamide and SBRT in our study was mild to moderate. The results of our study failed to meet the feasibility cutoff of 11 patients completing treat-ment, which formed part of the primary end point of the study: 10 of 15 patients received all 6 AMP-224 infusions. Disease progression was responsible for treatment noncompletion in the remaining patients. There was no difference in response or toxicity seen between the 2 arms. Because of the small patient population, it is difficult to determine if the radiation dosing schedule affected the changes in immune response seen in the tumor samples.
No objective response was observed, although disease stabiliza-tion was documented in 3 patients (3/15, 20%). In this heavily pretreated population of patients, the median PFS (mPFS) and OS were 2.8 months and 6.0 months, respectively. These results are comparable with previous placebo-controlled studies of third-line treatment in mCRC. Specifically, regorafenib resulted in a mPFS of 2.0 months in the CORRECT trial7 and 3.2 months in the CONCUR trial,43 whereas the TAS-102 resulted in a mPFS of 2.0
Table 1 Baseline Characteristics of Participants (n [ 15)
Metastasectomy 12 (80)
Number of Previous Regimens
Data are presented as n (%) or median (range).
Abbreviations: A ¼ Asian; B ¼ black; ECOG PS ¼ Eastern Cooperative Oncology Group performance status; M1 ¼ metastasis; MSS ¼ microsatellite stable; MT ¼ mutated; W ¼ white; WT ¼ wild type.
months.8 The placebo arm in all 3 studies was uniformly reported as having a median PFS of 1.7 months. In terms of OS, for the active treatment arms the median OS was 6.4 months in the CORRECT trial, 8.8 months in the CONCUR trial, and 7.1 months in the TAS-102 trial.7,8,43 The validity of such a comparison, however, is limited by the differences in the characteristics of the treated pa-tients, such as the number of previous regimens received by the
Table 2 Treatment-Related Adverse Events
Adverse Event Grade 1 Grade 2 Grade 3 Grade 4 Dizziness 1 Hypokalemia 1
Hypothyroidism 1 Infusion-Related Reaction 2 6
Lymphocyte Count Increased 1 Nausea 1
White Blood Cell Count 1 Decreased
Patients are counted once for each applicable specific adverse event and could have more than 1 treatment-related event. Charalampos S. Floudas et al
aWilson binomial confidence intervals.
patients. Meanwhile, AMP-224 targets to PD-L2 and executes in-hibition of PD-1/PD-L2 binding. It is not clear if the blockade of PD-1/PD-L2 by antiePD-L2 is not striking enough to inhibit the PD-1 signaling pathway to reflect clinical benefit compared with antiePD-L1. However, there is evidence Reading fram indicates the molec-