Tumor immune microenvironment of primary colorectal adenocarcinomas metastasizing to the liver or lungs.

BACKGROUND: Because of the heterogeneity of metastatic colorectal cancer (mCRC), a genome-wide analysis was performed to characterize the tumor immune microenvironment (TIME). METHODS: RNA-seq analysis of 62 primary CRCs without and 63 with systemic metastasis (SM- and SM+ groups) was conducted, and the data were used in a training set after adjustment by propensity score matching. Samples were further subdivided into those with hepatic metastasis (CHM subgroup), pulmonary metastasis (CPM subgroup), or concurrent CHM and CPM (concurrent group). Validation was done by quantitative reverse-transcription polymerase chain reaction using another 40 primary CRC samples. RESULTS: Compared with the CHM or CPM subgroups, the concurrent group showed upregulated in inflammatory or immune processes, cytokine secretion, and myeloid leukocyte migration. Nine candidate genes were selected: SM-specific IDO1, JAM3, and PDE2A; CHM- or CPM-specific BIRC7; CPM-specific HISI1H2BK, and both SM-specific and CHM- or CPM-specific EPHB6, LPL, THBD, and PPBP. In a validation set of primary CRCs, JAM3 and IDO1 (p = 0.044 and p = 0.036, respectively) were confirmed to show significant upregulation and downregulation, respectively, in the SM+ group, whereas HIST1H2BK (p = 0.017) was significantly upregulated in the CPM subgroup. CONCLUSIONS: Our findings indicate that a host-suppressive TIME is established in the primary tumor of mCRC and identify immune-related site-specific markers of mCRC.

Epigenetic regulation of KLHL34 predictive of pathologic response to preoperative chemoradiation therapy in rectal cancer patients.

PURPOSE: Prediction of individual responsiveness to preoperative chemoradiation therapy (CRT) is urgently needed in patients with poorly responsive locally advanced rectal cancer (LARC). METHODS AND MATERIALS: Candidate methylation genes associated with radiosensitivity were identified using a 3-step process. In the first step, genome-wide screening of methylation genes was performed in correlation with histopathologic tumor regression grade in 45 patients with LARC. In the second step, the methylation status of selected sites was analyzed by pyrosequencing in 67 LARC patients, including 24 patients analyzed in the first step. Finally, colorectal cancer cell clones with stable KLHL34 knockdown were generated and tested for cellular sensitivity to radiation. RESULTS: Genome-wide screening identified 7 hypermethylated CpG sites (DZIP1 cg24107021, DZIP1 cg26886381, ZEB1 cg04430381, DKK3 cg041006961, STL cg00991794, KLHL34 cg01828474, and ARHGAP6 cg07828380) associated with preoperative CRT responses. Radiosensitivity in patients with hypermethylated KLHL34 cg14232291 was confirmed by pyrosequencing in additional cohorts. Knockdown of KLHL34 significantly reduced colony formation (KLHL34 sh#1: 20.1%, P=.0001 and KLHL34 sh#2: 15.8%, P=.0002), increased the cytotoxicity (KLHL34 sh#1: 14.8%, P=.019 and KLHL34 sh#2: 17.9%, P=.007) in LoVo cells, and increased radiation-induced caspase-3 activity and the sub-G1 population of cells. CONCLUSIONS: The methylation status of KLHL34 cg14232291 may be a predictive candidate of sensitivity to preoperative CRT, although further validation is needed in large cohorts using various cell types.

Novel single-nucleotide polymorphism markers predictive of pathologic response to preoperative chemoradiation therapy in rectal cancer patients.

PURPOSE: Studies aimed at predicting individual responsiveness to preoperative chemoradiation therapy (CRT) are urgently needed, especially considering the risks associated with poorly responsive patients. METHODS AND MATERIALS: A 3-step strategy for the determination of CRT sensitivity is proposed based on (1) the screening of a human genome-wide single-nucleotide polymorphism (SNP) array in correlation with histopathologic tumor regression grade (TRG); (2) clinical association analysis of 113 patients treated with preoperative CRT; and (3) a cell-based functional assay for biological validation. RESULTS: Genome-wide screening identified 9 SNPs associated with preoperative CRT responses. Positive responses (TRG 1-3) were obtained more frequently in patients carrying the reference allele (C) of the SNP CORO2A rs1985859 than in those with the substitution allele (T) (P=.01). Downregulation of CORO2A was significantly associated with reduced early apoptosis by 27% (P=.048) and 39% (P=.023) in RKO and COLO320DM colorectal cancer cells, respectively, as determined by flow cytometry. Reduced radiosensitivity was confirmed by colony-forming assays in the 2 colorectal cancer cells (P=.034 and .015, respectively). The SNP FAM101A rs7955740 was not associated with radiosensitivity in the clinical association analysis. However, downregulation of FAM101A significantly reduced early apoptosis by 29% in RKO cells (P=.047), and it enhanced colony formation in RKO cells (P=.001) and COLO320DM cells (P=.002). CONCLUSION: CRT-sensitive SNP markers were identified using a novel 3-step process. The candidate marker CORO2A rs1985859 and the putative marker FAM101A rs7955740 may be of value for the prediction of radiosensitivity to preoperative CRT, although further validation is needed in large cohorts.

Genome-wide identification of possible methylation markers chemosensitive to targeted regimens in colorectal cancers.

PURPOSE: Few efficient methylation markers of chemosensitivity have been discovered. The genome-wide analysis of methylation markers is needed to identify chemosensitive candidates to targeted therapy. METHODS: This study describes a two-step process to select chemosensitive candidates of methylation genes. A genome-wide screening of methylation genes was performed using a Beadarray and an in vitro chemosensitivity assay of 119 colorectal cancers (CRCs). Ten candidate genes identified during the initial screening were verified by biological utility assessment using cell viability assays of transfected CRC cells. RESULTS: Five methylation genes related to sensitivity to bevacizumab regimens (RASSF1, MMP25, KCNQ1, ESR1, and GALR2) or cetuximab regimens (SCL18A2, GPX7, NID2, IGFBP3, and ALX4) were chosen during the first step. A viability assay revealed that GALR2-overexpressing HCT116 cells were significantly more chemosensitive to bevacizumab regimens than control cells (P = 0.022 and 0.019 for bevacizumab with FOLFIRI and FOLFOX, respectively), concurrently verified on a caspase-3 activity assay. GPX7- or ALX4-overexpressed RKO cells were significantly less viable to cetuximab regimens compared to control cells (GPX7: P = 0.027 each for cetuximab with FOLFIRI and FOLFOX; ALX4: P = 0.049 and 0.003 for cetuximab with FOLFIRI and FOLFOX, respectively), but caspase-3 activity was not prominent in GPX7-overexpressed RKO cells. CONCLUSIONS: Two novel genes, GALR2 and ALX4, have been identified as chemosensitive methylation candidates to bevacizumab and cetuximab regimens, respectively. As our study did not include a clinical association study, the two candidates should be validated in large clinical cohorts, hopefully predicting responsive patients to targeted regimens.

Novel chemosensitive single-nucleotide polymorphism markers to targeted regimens in metastatic colorectal cancer.

PURPOSE: Methods for predicting individual responsiveness to targeted chemotherapy are urgently needed, considering the frequent resistance and extremely high cost. EXPERIMENTAL DESIGN: A chemosensitive single-nucleotide polymorphism (SNP) discovery schema is presented that utilizes (i) genome-wide SNP screening with a human SNP array and an in vitro chemosensitivity assay in 118 colorectal cancers, (ii) clinical association analysis in the other 98 patients who had received chemotherapy for metastatic cancer, and (iii) biological utility assessment using cell viability assays of transfected colorectal cancer (CRC) cells. RESULTS: Nine SNPs related to bevacizumab and cetuximab regimen sensitivity were chosen during screening. Overall responses for bevacizumab regimens revealed that patients carrying the TT genotype at ANXA11 rs1049550 or at least one G allele at LINS1 rs11247226 seemed greater chemosensitive than those carrying at least one C allele or the AA genotype, respectively (P < 0.05). For cetuximab regimens, patients carrying the GG genotype at DFNB31 rs2274159 or LIFR rs3729740 seemed greater chemosensitive than those carrying at least one A allele (P = 0.025 and P = 0.07). Cytotoxicity analyses showed that all RKO and HCT116 CRC clones transfected with the G allele at LIFR rs3729740 and the C allele at ISX rs361863 were more sensitive to cetuximab regimens than those with the A and T allele, respectively (P ≤ 0.001-0.024). CONCLUSIONS: Chemosensitive SNP markers were identified using a novel three-step process. The candidate marker LIFR rs3729740 and possibly ISX rs361863 will hopefully predict responsive patients to cetuximab regimens, although further validation is needed in large cohorts.