Comprehensive prognostic analysis in breast cancer integrating clinical, tumoral, micro-environmental and immunohistochemical criteria

Machine Learning and Mechanistic Modeling for Prediction of Metastatic Relapse in Early-Stage Breast Cancer

PURPOSE

For patients with early-stage breast cancer, predicting the risk of metastatic relapse is of crucial importance. Existing predictive models rely on agnostic survival analysis statistical tools (eg, Cox regression). Here we define and evaluate the predictive ability of a mechanistic model for time to distant metastatic relapse.

METHODS

The data we used for our model consisted of 642 patients with 21 clinicopathologic variables. A mechanistic model was developed on the basis of two intrinsic mechanisms of metastatic progression: growth (parameter α) and dissemination (parameter μ). Population statistical distributions of the parameters were inferred using mixed-effects modeling. A random survival forest analysis was used to select a minimal set of five covariates with the best predictive power. These were further considered to individually predict the model parameters by using a backward selection approach. Predictive performances were compared with classic Cox regression and machine learning algorithms.

RESULTS

The mechanistic model was able to accurately fit the data. Covariate analysis revealed statistically significant association of Ki67 expression with α (P = .001) and EGFR expression with μ (P = .009). The model achieved a c-index of 0.65 (95% CI, 0.60 to 0.71) in cross-validation and had predictive performance similar to that of random survival forest (95% CI, 0.66 to 0.69) and Cox regression (95% CI, 0.62 to 0.67) as well as machine learning classification algorithms.

CONCLUSION

By providing informative estimates of the invisible metastatic burden at the time of diagnosis and forward simulations of metastatic growth, the proposed model could be used as a personalized prediction tool for routine management of patients with breast cancer.

FARP1 boosts CDC42 activity from integrin αvβ5 signaling and correlates with poor prognosis of advanced gastric cancer

Considering the poor prognosis of most advanced cancers, prevention of invasion and metastasis is essential for disease control. Ras homologous (Rho) guanine exchange factors (GEFs) and their signaling cascade could be potential therapeutic targets in advanced cancers. We conducted in silico analyses of The Cancer Genome Atlas expression data to identify candidate Rho-GEF genes showing aberrant expression in advanced gastric cancer and found FERM, Rho/ArhGEF, and pleckstrin domain protein 1 (FARP1) expression is related to poor prognosis. Analyses in 91 clinical advanced gastric cancers of the relationship of prognosis and pathological factors with immunohistochemical expression of FARP1 indicated that high expression of FARP1 is significantly associated with lymphatic invasion, lymph metastasis, and poor prognosis of the patients (P = 0.025). In gastric cancer cells, FARP1 knockdown decreased cell motility, whereas FARP1 overexpression promoted cell motility and filopodium formation via CDC42 activation. FARP1 interacted with integrin β5, and a potent integrin αvβ5 inhibitor (SB273005) prevented cell motility in only high FARP1-expressing gastric cancer cells. These results suggest that the integrin αvβ5-FARP1-CDC42 axis plays a crucial role in gastric cancer cell migration and invasion. Thus, regulatory cascade upstream of Rho can be a specific and promising target of advanced cancer treatment.

Design and analysis of nested case-control studies for recurrent events subject to a terminal event

The process by which patients experience a series of recurrent events, such as hospitalizations, may be subject to death. In cohort studies, one strategy for analyzing such data is to fit a joint frailty model for the intensities of the recurrent event and death, which estimates covariate effects on the two event types while accounting for their dependence. When certain covariates are difficult to obtain, however, researchers may only have the resources to subsample patients on whom to collect complete data: one way is using the nested case-control (NCC) design, in which risk set sampling is performed based on a single outcome. We develop a general framework for the design of NCC studies in the presence of recurrent and terminal events and propose estimation and inference for a joint frailty model for recurrence and death using data arising from such studies. We propose a maximum weighted penalized likelihood approach using flexible spline models for the baseline intensity functions. Two standard error estimators are proposed: a sandwich estimator and a perturbation resampling procedure. We investigate operating characteristics of our estimators as well as design considerations via a simulation study and illustrate our methods using two studies: one on recurrent cardiac hospitalizations in patients with heart failure and the other on local recurrence and metastasis in patients with breast cancer.

Ras protein expression as a marker for breast cancer

Breast cancer, the most common neoplasm in women of all ages, is the leading cause of cancer-related mortality in women worldwide. Markers to help to predict the risk of progression and ultimately provide non-surgical treatment options would be of great benefit. At present, there are no available molecular markers to predict the risk of carcinoma in situ progression to invasive cancer; therefore, all women diagnosed with this type of malignancy must undergo surgery. Breast cancer is a heterogeneous complex disease, and different patients respond differently to different treatments. In breast cancer, analysis using immunohistochemical markers remains an essential component of routine pathological examinations, and plays an import role in the management of the disease by providing diagnostic and prognostic strategies. The aim of the present study was to identify a marker that can be used as a prognostic tool for breast cancer. For this purpose, we firstly used an established breast cancer model. MCF-10F, a spontaneously immortalized breast epithelial cell line was transformed by exposure to estrogen and radiation. MCF-10F cells were exposed to low doses of high linear energy transfer (LET) α particles (150 keV/μm) of radiation, and subsequently cultured in the presence of 17β-estradiol. Three cell lines were used: i) MCF-10F cells as a control; ii) Alpha5 cells, a malignant and tumorigenic cell line; and iii) Tumor2 cells derived from Alpha5 cells injected into nude mice. Secondly, we also used normal, benign and malignant breast specimens obtained from biopsies. The results revealed that the MCF-10F cells were negative for c-Ha-Ras protein expression; however, the Alpha5 and Tumor2 cell lines were positive for c-Ha-Ras protein expression. The malignant breast samples were also strongly positive for c-Ha-Ras expression. The findings of our study indicate that c-Ha-Ras protein expression may be used as a marker to predict the progression of breast cancer; this marker may also ultimately provide non-surgical treatment options for patients who are at a lower risk.