Miguel Henriques Abreu

Missing data imputation on the 5-year survival prediction of breast cancer patients with unknown discrete values

Breast cancer is the most frequently diagnosed cancer in women. Using historical patient information stored in clinical datasets, data mining and machine learning approaches can be applied to predict the survival of breast cancer patients. A common drawback is the absence of information, i.e., missing data, in certain clinical trials. However, most standard prediction methods are not able to handle incomplete samples and, then, missing data imputation is a widely applied approach for solving this inconvenience. Therefore, and taking into account the characteristics of each breast cancer dataset, it is required to perform a detailed analysis to determine the most appropriate imputation and prediction methods in each clinical environment. This research work analyzes a real breast cancer dataset from Institute Portuguese of Oncology of Porto with a high percentage of unknown categorical information (most clinical data of the patients are incomplete), which is a challenge in terms of complexity. Four scenarios are evaluated: (I) 5-year survival prediction without imputation and 5-year survival prediction from cleaned dataset with (II) Mode imputation, (III) Expectation-Maximization imputation and (IV) K-Nearest Neighbors imputation. Prediction models for breast cancer survivability are constructed using four different methods: K-Nearest Neighbors, Classification Trees, Logistic Regression and Support Vector Machines. Experiments are performed in a nested ten-fold cross-validation procedure and, according to the obtained results, the best results are provided by the K-Nearest Neighbors algorithm: more than 81% of accuracy and more than 0.78 of area under the Receiver Operator Characteristic curve, which constitutes very good results in this complex scenario.

Male breast cancer: Looking for better prognostic subgroups.

PURPOSE Male Breast Cancer (MBC) remains a poor understood disease. Prognostic factors are not well established and specific prognostic subgroups are warranted. PATIENTS/METHODS Retrospectively revision of 111 cases treated in the same Cancer Center. Blinded-central pathological revision with immunohistochemical (IHQ) analysis for estrogen (ER), progesterone (PR) and androgen (AR) receptors, HER2, ki67 and p53 was done. Cox regression model was used for uni/multivariate survival analysis. Two classifications of Female Breast Cancer (FBC) subgroups (based in ER, PR, HER2, 2000 classification, and in ER, PR, HER2, ki67, 2013 classification) were used to achieve their prognostic value in MBC patients. Hierarchical clustering was performed to define subgroups based on the six-IHQ panel. RESULTS According to FBC classifications, the majority of tumors were luminal: A (89.2%; 60.0%) and B (7.2%; 35.8%). Triple negative phenotype was infrequent (2.7%; 3.2%) and HER2 enriched, non-luminal, was rare (≤1% in both). In multivariate analysis the poor prognostic factors were: size >2 cm (HR:1.8; 95%CI:1.0-3.4 years, p = 0.049), absence of ER (HR:4.9; 95%CI:1.7-14.3 years, p = 0.004) and presence of distant metastasis (HR:5.3; 95%CI:2.2-3.1 years, p < 0.001). FBC subtypes were independent prognostic factors (p = 0.009, p = 0.046), but when analyzed only luminal groups, prognosis did not differ regardless the classification used (p > 0.20). Clustering defined different subgroups, that have prognostic value in multivariate analysis (p = 0.005), with better survival in ER/PR+, AR-, HER2-and ki67/p53 low group (median: 11.5 years; 95%CI: 6.2-16.8 years) and worst in PR-group (median:4.5 years; 95%CI: 1.6-7.8 years). CONCLUSION FBC subtypes do not give the same prognostic information in MBC even in luminal groups. Two subgroups with distinct prognosis were identified in a common six-IHQ panel. Future studies must achieve their real prognostic value in these patients.

Predicting Breast Cancer Recurrence Using Machine Learning Techniques: A Systematic Review

Background: Recurrence is an important cornerstone in breast cancer behavior, intrinsically related to mortality. In spite of its relevance, it is rarely recorded in the majority of breast cancer datasets, which makes research in its prediction more difficult. Objectives: To evaluate the performance of machine learning techniques applied to the prediction of breast cancer recurrence. Material and Methods: Revision of published works that used machine learning techniques in local and open source databases between 1997 and 2014. Results: The revision showed that it is difficult to obtain a representative dataset for breast cancer recurrence and there is no consensus on the best set of predictors for this disease. High accuracy results are often achieved, yet compromising sensitivity. The missing data and class imbalance problems are rarely addressed and most often the chosen performance metrics are inappropriate for the context. Discussion and Conclusions: Although different techniques have been used, prediction of breast cancer recurrence is still an open problem. The combination of different machine learning techniques, along with the definition of standard predictors for breast cancer recurrence seem to be the main future directions to obtain better results.

Overall Survival Prediction for Women Breast Cancer Using Ensemble Methods and Incomplete Clinical Data

Breast Cancer is the most common type of cancer in women worldwide. In spite of this fact, there are insufficient studies that, using data mining techniques, are capable of helping medical doctors in their daily practice.

CYP2D6*4 polymorphism: A new marker of response to hormonotherapy in male breast cancer?

BACKGROUND Tamoxifen remains the standard hormonotherapy for Male breast cancer patients (MBC). Previous studies, in women, tried to evaluate the impact of CYP2D6 polymorphisms in tamoxifen efficacy with conflicting results. Herein we analyze the relation between CYP2D6*4 polymorphism and survival in MBC patients. PATIENTS AND METHODS Fifty-three patients, proposed to tamoxifen in adjuvant setting, were enrolled. Clinical information was collected from records and histological revision with additional immunochemistry analysis was done to better characterize the tumors. Comprehensive CYP2D6*4 genotyping from blood or tumor tissue was performed and translated into two predicted metabolic activity groups. RESULTS Patients included in the two CYP2D6*4 groups did not differ concerning to age, histological characteristics, and primary treatments performed. Median age at diagnosis was 63 years-old and patients were submitted at least to mastectomy and adjuvant hormonotherapy. Recurrence was observed in 7 patients (13.2%) and 13 patients (25.5%) died with a 5-year disease-free survival of 86.2%. The poorer metabolizer group had a high risk for recurrence (p = 0.034) and this outcome effect remains in different subgroups: in tumors larger than 2 cm (p < 0.001), nodal status, N0 vs N+ (p = 0.04) and in advanced stage, stage III (p < 0.001). Poorer metabolizer patients had also a worse overall survival when tumors were larger than 2 cm (p = 0.03). CONCLUSIONS In our series, there was an association between CYP2D6*4 polymorphism and a probability of recurrence, with a consistent effect in risk groups defined by classic prognostic factors. Multicentric studies with larger samples are needed to validate these results.

Personalizing Breast Cancer Patients with Heterogeneous Data

The prediction of overall survival in patients has an important role, especially in diseases with a high mortality rate. Encompassed in this reality, patients with oncological diseases, particularly the more frequent ones like woman breast cancer, can take advantage of a very good customization, which in some cases may even lead to a disease-free life. In order to achieve this customization, in this work a comparison between three algorithms (evolutionary, hierarchical and k-medoids) is proposed. After constructing a database with more than 800 breast cancer patients from a single oncology center with 15 clinical variables (heterogeneous data) and having 25% of the data missing, which illustrates a real clinical scenario, the algorithms were used to group similar patients into clusters. Using Tukey’s HSD (Honestly Significant Difference) test, from both comparison between k-medoids and the other two approaches (evolutionary and hierarchical clustering) a statistical difference were detected (p − value < 0.0000001) as well as for the other comparison (evolutionary versus hierarchical clustering) - p − value = 0.0061354 - for a significance level of 95%.

Staging and survival of rectal cancer in Vila Nova de Gaia, Portugal.

Introduction In the county of Vila Nova de Gaia (northern Portugal) in the period of 2004–2006, there were an average of 35 new cases of colorectal cancer per 100 000 population, which constitutes one of the highest rates in the world. The latest research has shown that there are many differences between colon and rectal cancers, thereby justifying an independent approach. Patients and methods The study pertained to the period 1995–2004, by using the census of 1991 and 2001 for calculating specific rates. The 399 diagnosed cases of rectal cancer were drawn from a specialized and active cancer registry, oncological registry of Gaia. Overall survival was calculated using the Kaplan–Meier method and the curves were compared using a Log Rank test. The effect of topography and histological type on survival was obtained by controlling the stage disease, using a Cox proportional hazards regression model. Results There was a slight predominance of males, with a ratio between sexes of 1 : 3. The 50% overall survival rate after 5 years increased over time. The localization of the tumour and the histological type, after adjusting by stage, were not significant factors in the prognosis. Conclusion Our study shows an increase in the number of cases over time, particularly in elderly women. The cumulative risk of having rectal cancer remains unchanged from 1981 to 2004. Unlike other studies, an increase in early lesions was not observed.

Rectal Cancer Epidemiology

Colorectal cancer is the fourth most common cancer in men and the third most common one in women worldwide (Parkin, 2004; Parkin et al., 2005), accounting for approximately 436,000 incident cases and 212,000 deaths in 2008 (Quirke et al., 2011). This cancer has an important economic impact, estimating that in the initial, continuing and last year of life phases of care a total of more than

Patterns of recurrence and treatment in male breast cancer: A clue to prognosis?

7 billion were spent (Yabroff et al., 2008). Randomized trials have shown that systematic screening of a target population of suitable age can reduce colorectal cancer by detecting asymptomatic lesions (Center et al., 2009). Although there are differences in the etiologies and epidemiology of colon and rectal cancer (Giovannucci & Wu, 2006), the majority of the studies chose to examine colon and rectum cancers combined. However, a better understanding of these diseases nowadays, shows that these differences have an important impact in their approaches. First of all, the location of the tumours may determines different locations of metastisation. Unlike colon cancers, distal rectal tumours may first metastasize to the lungs because the inferior rectal veins drain into the inferior vena cava rather than into the portal venous system. The histological type can also vary. The vast majority of colorectal tumours are adenocarcinomas but 11-17% are mucinous carcinomas. This type, which has a penchant for the rectum and sigmoid colon, tends to be present at a more advanced stage (Consorti et al., 2000). The carcinoid tumours have a different clinical presentation too, depending on whether they appear in the rectum or in the colon (Marshall & Badnarchuk 1993; Spread et al., 1994). The rectum carcinoids develop at a young age, most of which are less than 2 cm and tend to be indolent. In contrast, colonic carcinoid tumours can be clinically aggressive and often metastise. With a more accurate review, we can see that many habits could influence the development of rectal cancers and not colon cancers. Some studies support the view that family history, as well as the level of physical activity, is a stronger contributor to colon cancer relative to rectal cancer (Wei et al., 2004). The Women’s Health Initiative (a large cohort study) (Paskett et al., 2007) also found a significant link between active cigarette smoking (not passive exposure to cigarette smoke) and rectal but not colon cancer. These differences are important in terms of monitoring and have implications in treatment options, as well. Compared to colon cancers, the sensitivity of CT scan for detection of