Risk factors affecting patients survival with colorectal cancer in Morocco: survival analysis using an interpretable machine learning approach.

The aim of our study was to assess the overall survival rates for colorectal cancer at 3 years and to identify associated strong prognostic factors among patients in Morocco through an interpretable machine learning approach. This approach is based on a fully non-parametric survival random forest (RSF), incorporating variable importance and partial dependence effects. The data was povided from a retrospective study of 343 patients diagnosed and followed at Hassan II University Hospital. Covariate selection was performed using the variable importance based on permutation and partial dependence plots were displayed to explore in depth the relationship between the estimated partial effect of a given predictor and survival rates. The predictive performance was measured by two metrics, the Concordance Index (C-index) and the Brier Score (BS). Overall survival rates at 1, 2 and 3 years were, respectively, 87% (SE = 0.02; CI-95% 0.84-0.91), 77% (SE = 0.02; CI-95% 0.73-0.82) and 60% (SE = 0.03; CI-95% 0.54-0.66). In the Cox model after adjustment for all covariates, sex, tumor differentiation had no significant effect on prognosis, but rather tumor site had a significant effect. The variable importance obtained from RSF strengthens that surgery, stage, insurance, residency, and age were the most important prognostic factors. The discriminative capacity of the Cox PH and RSF was, respectively, 0.771 and 0.798 for the C-index while the accuracy of the Cox PH and RSF was, respectively, 0.257 and 0.207 for the BS. This shows that RSF had both better discriminative capacity and predictive accuracy. Our results show that patients who are older than 70, living in rural areas, without health insurance, at a distant stage and who have not had surgery constitute a subgroup of patients with poor prognosis.

Occupation and risk of female breast cancer: A case-control study in Morocco.

BACKGROUND: Breast cancer is the most frequent cancer among Moroccan women. Environmental and occupational factors may play a role in breast cancer etiology. This study aimed to investigate the association between occupation, industry, and breast cancer risk among Moroccan women. METHODS: A total of 300 breast cancer cases and 300 controls (matched by age and area of residence) were included in this study. Full occupational history was collected, with a detailed description of each job held for at least 6 months. Occupations were coded according to the International Standard Classification of Occupations (ISCO 08) and the Moroccan Analytical Classification of Professions (2001). Industries were coded according to the Statistical Classification of Economic Activities in the European Community (2008). Odds ratios (ORs) and 95% confidence intervals (CIs), adjusted for potential confounders were estimated by using conditional logistic regression. RESULTS: An overall decreased risk of breast cancer was observed among women doing only household work (OR = 0.32; 95% CI = 0.18-0.55). An increased risk of breast cancer was observed among women in agricultural occupations, particularly those employed as agricultural laborers (ISCO 08 code: 921; OR = 2.91; 95% CI = 1.51-5.60) and the risk increased with duration of employment (P trend = .01). Analyses by industry corroborated these findings. CONCLUSIONS: Our findings suggest that occupational exposures may be associated with increased risk of breast cancer among female agricultural workers in this population. Further investigations, with advanced methods of occupational exposure assessment, are warranted to clarify the role of chemicals involved in this high-risk occupation and to suggest preventive actions and screening.

Prevalence of molecular subtypes and prognosis of invasive breast cancer in north-east of Morocco: retrospective study.

BACKGROUND: Breast carcinoma is known as a heterogeneous disease because gene expression analyses identify several subtypes and the molecular profiles are prognostic and predictive for patients. Our aim, in this study, is to estimate the prevalence of breast cancer subtypes and to determine the relationship between clinico-pathological characteristics, overall survival (OS) and disease free survival (DFS) for patients coming from north-east of Morocco. METHODS: We reviewed 366 cases of breast cancer diagnosed between January 2007 to June 2010 at the Department of pathology. Age, size tumor, metastatic profile, node involvement profile, OS and DFS were analyzed on 181 patients. These last parameters were estimated by Kaplan-Meier analysis and log-rank test to estimate outcome differences among subgroups. RESULTS: The average age was 45 years, our patients were diagnosed late (57% stage III, 17.5% stage IV) with a high average tumor size. Luminal A subtype was more prevalent (53.6%) associated with favorable clinic-pathological characteristics, followed by luminal B (16.4%), Her2-overexpressing (12.6%), basal-like (12.6%) and unclassified subtype (4.9%).Survival analysis showed a significant difference between subtypes. The triple negative tumors were associated with poor prognosis (49% OS, 39% DFS), whereas the luminal A were associated with a better prognosis (88% OS, 59% DFS). The luminal B and the Her2-overexpressing subtypes were associated with an intermediate prognosis (77% and 75% OS, and 41% and 38% DFS respectively). CONCLUSION: This study showed that molecular classification by immunohistochemistry was necessary for therapeutic decision and prognosis of breast carcinoma. The luminal A subtype was associated with favorable biological characteristics and a better prognosis than triple negative tumors that were associated with a poor prognosis and unfavorable clinic-pathological characteristics.

Clinicopathological, therapeutic and prognostic features of the triple-negative tumors in moroccan breast cancer patients (experience of Hassan II university hospital in Fez).

INTRODUCTION: Triple-negative breast cancer (TNBC) is defined as a group of breast carcinomas that are negative for expression of hormone receptors (ER, PR) and Her2, we can distinguish between two groups: basal-like (ER-, PR-, Her2-, cytokeratin (CK) 5/6+ and/or Her1+) and unclassified subtype (ER-, PR-, Her2-, Her1- and CK5/6-).The aim of this study is to determine the clinicopathological, histological, therapeutic and prognostic features associated with this type of breast cancer. METHODS: This is a retrospective study of 366 female breast cancer patients, diagnosed between January 2007 and June 2010 at the Department of Pathology. Epidemiological, clinical, histological, therapeutic and evolutive data were analyzed. OS and DFS rates were estimated by Kaplan-Meier analysis and a log-rank test to estimate outcome. RESULTS: A total of 64 women were identified as having TNBC (17.5% of all female breast cancer patients), 12.6% were basal-like, 4.9% were unclassified subtype, with a median age of 45 years. The median histological tumor diameter was 4.3 cm. TNBC were most often associated with a high grade, 49.2% grade III (53% for unclassified subtype, 47.6% for basal-like). Vascular invasion was found in 26.6% of cases (22% for unclassified subtype and 28.3% for basal-like). For the lymph node involvement: 51% had positive lymph nodes, and 22.4% had distant metastases. Neoadjuvant chemotherapy was administered to 18% patients with 26% of complete pathologic response; therefore adjuvant chemotherapy was given to 82%. 98% received anthracycline based regimen and only 30% received taxanes.The Kaplan-Meier curves based showed the lowest survival probability at 3-years (49% of OS, and 39% of DFS). CONCLUSION: TNBC is associated with young age, high grade tumors, advanced stage at diagnosis, difference chemo response compared to other subtypes, and shortest survival. Critical to optimal future management is accurate identification of truly triple negative disease, and adequately powered prospective TNBC trials to establish treatment efficacy and define predictive biomarkers.