RESUMO
BACKGROUND: Patients with T4 colon adenocarcinomas have an increased risk of peritoneal metastases (PM) but the histopathologic risk factors for its development are not well-described. OBJECTIVE: The purpose of this study was to determine factors associated with PM, time to recurrence, and survival after recurrence among patients with T4 colon cancer. PATIENTS AND METHODS: Patients with pathologic T4 colon cancer who underwent curative resection from 2005 to 2017 were identified from a prospectively maintained institutional database and classified by recurrence pattern: (a) none - 68.8%; (b) peritoneal only - 7.9%; (c) peritoneal and extraperitoneal - 9.9%; and (d) extraperitoneal only - 13.2%. Associations between PM development and patient, primary tumor, and treatment factors were assessed. RESULTS: Overall, 151 patients were analyzed, with a median follow-up of 66.2 months; 27 patients (18%) developed PM (Groups B and C) and 20 (13%) patients recurred at non-peritoneal sites only (Group D). Median time to developing metastases was shorter for Groups B and C compared with Group D (B and C: 13.7 months; D: 46.7 months; p = 0.022). Tumor deposits (TDs) and nodal stage were associated with PM (p < 0.05), and TDs (p = 0.048) and LVI (p = 0.015) were associated with additional extraperitoneal recurrence. Eleven (41%) patients with PM underwent salvage surgery, and median survival after recurrence was associated with the ability to undergo cytoreduction (risk ratio 0.20, confidence interval 0.06-0.70). CONCLUSION: PM risk after resection of T4 colon cancer is independently associated with factors related to lymphatic spread, such as N stage and TDs. Well-selected patients can undergo cytoreduction with long-term survival. These findings support frequent postoperative surveillance and aggressive early intervention, including cytoreduction.
RESUMO
BACKGROUND: The clinical utility of a blood-based biomarker in squamous cell carcinoma of the anus (SCCA) is unknown. We analyzed carcinoembryonic antigen (CEA), a commonly employed assay for patients with colorectal adenocarcinoma, as a serum biomarker for patients with biopsy-proven SCCA. MATERIALS AND METHODS: Medical records from 219 patients with biopsy-proven SCCA at the University of Texas MD Anderson Cancer Center were reviewed under an IRB-approved protocol from 2013 to 2020 to assess for correlations between CEA levels and corresponding clinical and pathologic characteristics. RESULTS: The mean CEA among subgroups by clinical status at the time of presentation to our institution was highest among those patients with metastatic SCCA to visceral organs (M-V, 20.7 ng/mL), however this finding was not statistically significant by ANOVA (p = .74). By clinical subgroup, the percentage of patients with an abnormally elevated CEA was highest in those patients with metastatic disease to lymph nodes (M-L, 41.2%) followed by recurrent/unresectable SCCA (36.8%), and metastatic SCCA to visceral organs (M-V, 35.2%), and was statistically significant between groups (Fisher's exact test p = .02). Using RECIST criteria for tumor progression and disease response, the mean change in CEA for patients with progression was an increase in 19 ng/mL, compared to a change of -7.3 ng/mL in those with disease response (p = .004). We likewise assessed whether CEA levels were associated with survival outcomes for all patients with metastatic SCCA, and found no correlation between CEA and likelihood for survival in a ROC analysis (multivariate, age-adjusted analysis for CEA cutoff of 8, HR = 1.01, 95% CI 0.52-1.96). CONCLUSIONS: Despite interesting patterns of abnormally high CEA in SCCA patients with advanced disease, and correlation of increased CEA with disease progression (and conversely decreased CEA with disease response), CEA is not associated with survival outcomes in SCCA, and is not a clinically relevant biomarker in this disease.
RESUMO
Standard treatment for locally advanced rectal cancer (LARC) is neoadjuvant chemoradiotherapy (NACRT), followed by surgical resection. However, >70% of patients do not achieve a complete pathological response and have higher rates of relapse and death. There are no validated pre- or on-treatment factors that predict response to NACRT besides tumour stage and size. We characterised the response of 33 LARC patients to NACRT, collected tumour samples from patients prior to, during and after NACRT, and performed whole exome, transcriptome and high-depth targeted sequencing. The pre-treatment LARC genome was not predictive of response to NACRT. However, in line with the increasing recognition of microbial influence in cancer, RNA analysis of pre-treatment tumours suggested a greater abundance of Fusobacteria in intermediate and poor responders. In addition, we investigated tumour heterogeneity and evolution in response to NACRT. While matched pre-treatment, on-treatment and post-treatment tumours revealed minimal genome evolution overall, we identified cases in which microsatellite instability developed or was selected for during NACRT. Recent research has suggested a role for adaptive mutability to targeted therapy in colorectal cancer cells. We provide preliminary evidence of selection for mismatch repair deficiency in response to NACRT. Furthermore, pre-NACRT genomic landscapes do not predict treatment response but pre-NACRT microbiome characteristics may be informative.