Response to Chemotherapy and Prognosis in Metastatic Colorectal Cancer With DNA Deficient Mismatch Repair.

BACKGROUND: DNA deficient mismatch repair (dMMR) genes are associated with microsatellite instability and good prognosis in early-stage colorectal cancer (CRC). However dMMR is rare in metastatic CRC (mCRC) and little is known about its influence on treatment response rate (RR). The primary objective of this study was to compare the RR of patients with mCRC according to dMMR status. METHODS: This was a retrospective study that compared the RR by Response Evaluation Criteria In Solid Tumors 1.1 criteria in patients with mCRC treated with chemotherapy according to dMMR status. All digital images were retrieved for RR evaluation by a single radiologist blinded to dMMR results. dMMR was defined as loss of immunohistochemistry expression of at least 1 of the MMR genes (MLH1, MSH2, MSH6, or PMS2). Cases were dMMR patients, and controls were proficient MMR (pMMR) patients (1:2 fashion). Based on clinical and molecular features, dMMR patients were classified as probable Lynch or sporadic. RESULTS: From January 2009 to January 2013, 762 out of 1270 patients were eligible and screened for dMMR: n = 27 (3.5%) had dMMR mCRC and n = 735 (96.5%) had pMMR mCRC. Given the rarity, 14 dMMR cases outside the inclusion period were included (total 41 dMMR cases) and 84 controls (pMMR). By intention-to-treat analysis, considering all patients who received at least 1 dose of oxaliplatin-based chemotherapy (N dMMR = 34), those with dMMR had lower RR compared with those with pMMR (RR, 11.7% vs. 28.6%; odds ratio, 0.33; 95% confidence interval, 0.08-1.40; P = .088); patients with probable Lynch-related mCRC presented higher RR than subjects with probable sporadic dMMR (22.2% vs. 0%). dMMR was associated with BRAF mutations and poor prognosis, particularly in the sporadic subgroup (median survival, 29.8 vs. 5.9 months; P = .025). CONCLUSION: This study suggests that the dMMR phenotype is predictive of resistance to oxaliplatin-based chemotherapy. Apparently, such resistance is more pronounced in the sporadic dMMR phenotype, suggesting biological heterogeneity within the dMMR mCRC subgroup.

A influência da instabilidade de microssatélites e outros biomarcadores nos desfechos clínicos de pacientes com câncer colorretal metastático: um estudo caso-controle / The influence of microsatellite instability and other biomarkers on the clinical outcomes of patients with metastatic colorectal cancer: a case-control study

O câncer colorretal metastático (CCRm) é uma doença clinicamente e molecularmente heterogênea. Os pacientes apresentam diferentes prognósticos e respostas variáveis às terapias direcionadas contra o tumor. Alterações na função do sistema de reparo do DNA (deficiency mismatch repair - dMMR) estão associadas com o fenótipo de instabilidade de microssatélites e bom prognóstico em tumores de estádio inicial. No entanto, dMMR é raro no CCRm e pouco se sabe sobre sua influência na taxa de resposta (TR) ao tratamento. Nosso objetivo primário foi comparar a TR, de acordo com o status dMMR, nos pacientes com CCRm. Os desfechos secundários foram TR, conforme RAS e BRAF mutados, e a sobrevida global (SG), de acordo com dMMR. MÉTODOS: Estudo retrospectivo com grupo controle que comparou a TR por RECIST 1.1 em pacientes com CCRm, tratados com quimioterapia (QT) sistêmica, de acordo com o status dMMR. Os dados clínicos foram coletados, retrospectivamente, dos prontuários médicos. Todas as imagens foram digitais e recuperadas para avaliação de resposta por um único radiologista, cego quanto ao status dMMR. dMMR foi definido como a perda de expressão imuno-histoquímica em pelo menos um dos genes MMR (MLH1, MSH2, MSH6 e PMS2). Mutações em RAS e BRAF foram investigadas por meio de sequenciamento gênico. Os casos foram os pacientes com dMMR, e os controles, com MMR proficiente (pMMR), selecionados de forma consecutiva, em proporção de 1:2. Com base em características clínicas e moleculares, os indivíduos dMMR foram classificados como provável Lynch ou dMMR esporádico. Estatística descritiva foi usada para resumir os resultados. A associação entre dMMR e os resultados específicos de cada grupo foram analisados pelo teste do qui-quadrado, e para a avaliação de SG mediana, curvas de Kaplan-Meier e teste log-rank foram utilizados. Valores bicaudados de p < 0.05 foram considerados significativos. RESULTADOS: Entre janeiro de 2009 e janeiro de 2013, de...

Metastatic colorectal cancer (mCRC) is a clinically and molecularly heterogeneous disease, where patients present different prognosis and variable responses to cancer-directed therapies. Alterations in the function of DNA deficiency mismatch repair (dMMR) genes are associated with microsatellite instability and good prognosis in early stage tumors. However dMMR dysfunction is rare in mCRC and little is known about its influence on treatment response rate (RR). Our primary endpoint was to compare the RR of mCRC patients according to dMMR status and to explore differences between patients with likely sporadic versus likely Lynch-related tumors. Secondary endpoints were RR according to RAS and BRAF mutation status, and survival times as per dMMR status. METHODS: Retrospective study with control group that compared the RR by RECIST 1.1 in patients with mCRC treated with systemic chemotherapy according to dMMR status. Clinical data were collected retrospectively from medical charts. All images were digital and were retrieved for response evaluation by a single radiologist blinded to dMMR results. dMMR status was defined as loss of immunohistochemistry expression in at least one of the MMR genes (MLH1, MSH2, MSH6 e PMS2). RAS and BRAF mutations were investigated through next generation sequencing. Cases were defined as dMMR and controls, as proficient MMR (pMMR) patients, in a 1:2 fashion. Based on clinical and molecular features, dMMR patients were classified as likely Lynch or sporadic. Descriptive statistics was used to summarize the results. The association between dMMR and outcomes of each group were analyzed by chi-square test; estimates of median overall survival were done by the Kaplan-Meier method and comparisons, by the log-rank test. Two-tailed p values < 0.05 were considered significant. RESULTS: From January 2009 to January 2013, out of 1270 patients, 762 were eligible and screened for dMMR: N = 27 (3.5%) had dMMR and N = 735 (96.5%) had...

The effects of palliative chemotherapy in metastatic colorectal cancer patients with an ECOG performance status of 3 and 4.

BACKGROUND: Although chemotherapy is standard for patients with mCRC and ECOG PS of 0/1, the real benefit for patients with ECOG PS > 2 remains uncertain, because they are generally excluded from clinical trials. Our objectives were to compare the survival and safety of ECOG PS 3/4 patients who were administered chemotherapy with those who received BSC only. PATIENTS AND METHODS: We retrospectively analyzed all consecutive mCRC patients who started first-line chemotherapy at our institution in a 4-year period. A multivariable Cox regression model was used to adjust for prognostic factors and logistic regression, to identify predictive factors of Grade 3/4 toxicity. RESULTS: From June 2008 to June 2012, 240 consecutive patients were included: 100 (41.7%) had an ECOG PS of 0/1, 75 (31.3%) ECOG PS of 2, and 65 (27%) ECOG PS of 3/4. Median survival for patients treated with chemotherapy was 18.4 months for patients with ECOG PS of 0/1, 10.8 months for those with ECOG PS of 2, and 6.8 months for patients with ECOG PS of 3/4. Among those with ECOG PS of 3/4, chemotherapy use led to a nonsignificant survival gain (median, 6.8 vs. 2.3 months for BSC; P = .13). Factors significantly associated with worse survival in an adjusted analysis were right-sided tumors (hazard ratio [HR], 2.97; P = .005) and ECOG PS status (ECOG PS 2 vs. 0/1; HR, 1.67; P = .025, and ECOG PS 3/4 vs. 0/1; HR, 2.67; P < .0001). The rate of Grade ≥ 3 toxicities during the first cycle did not differ significantly across ECOG groups; likely because 40% of ECOG PS 3/4 patients received upfront dose-reduced therapy. The rates of treatment-related hospitalization were similar across all ECOG groups. All deaths were disease-associated. CONCLUSION: Our retrospective study suggests that chemotherapy might benefit selected mCRC patients with poor PS. With up-front dose reduction and close monitoring for toxicity, the risk of serious adverse events is minimized.