Impact of the enhanced permeability and retention (EPR) effect and cathepsins levels on the activity of polymer-drug conjugates

Polymer-drug conjugates have demonstrated clinical potential in the context of anticancer therapy. However, such promising results have, to date, failed to translate into a marketed product. Polymer-drug conjugates rely on two factors for activity: (i) the presence of a defective vasculature, for passive accumulation of this technology into the tumour tissue (enhanced permeability and retention (EPR) effect) and (ii) the presence of a specific trigger at the tumour site, for selective drug release (e.g., the enzyme cathepsin B). Here, we retrospectively analyse literature data to investigate which tumour types have proved more responsive to polymer-drug conjugates and to determine correlations between the magnitude of the EPR effect and/or expression of cathepsin B. Lung, breast and ovarian cancers showed the highest response rate (30%, 47% and 41%, respectively for cathepsin-activated conjugates and 31%, 43%, 40%, across all conjugates). An analysis of literature data on cathepsin content in various tumour types showed that these tumour types had high cathepsin content (up to 3835 ng/mg for lung cancer), although marked heterogeneity was observed across different studies. In addition, these tumour types were also reported as having a high EPR effect. Our results suggest that a pre-screening of patient population could bring a more marked clinical benefit.

Multiprobabilistic prediction in early medical diagnoses

This paper describes the methodology of providing multiprobability predictions for proteomic mass spectrometry data. The methodology is based on a newly developed machine learning framework called Venn machines. Is allows to output a valid probability interval. The methodology is designed for mass spectrometry data. For demonstrative purposes, we applied this methodology to MALDI-TOF data sets in order to predict the diagnosis of heart disease and early diagnoses of ovarian cancer and breast cancer. The experiments showed that probability intervals are narrow, that is, the output of the multiprobability predictor is similar to a single probability distribution. In addition, probability intervals produced for heart disease and ovarian cancer data were more accurate than the output of corresponding probability predictor. When Venn machines were forced to make point predictions, the accuracy of such predictions is for the most data better than the accuracy of the underlying algorithm that outputs single probability distribution of a label. Application of this methodology to MALDI-TOF data sets empirically demonstrates the validity. The accuracy of the proposed method on ovarian cancer data rises from 66.7 % 11 months in advance of the moment of diagnosis to up to 90.2 % at the moment of diagnosis. The same approach has been applied to heart disease data without time dependency, although the achieved accuracy was not as high (up to 69.9 %). The methodology allowed us to confirm mass spectrometry peaks previously identified as carrying statistically significant information for discrimination between controls and cases.