COMPLEX SYSTEMS FOR THE ECONOMIC EVALUATION OF HEALTHCARE IN ONCOLOGY

The health economy is witnessing a hard fight in recent years. On one side, the cut of resources by the governments and, on the other side, the increase in health demand. Additionally, another complex factor is represented by the rapid increase in the cost of new-generation drugs, especially in the oncology field, linked to the development of new molecular targets for immunotherapy. The purpose of this manuscript is to evaluate the relationship between the high costs of drugs emerging in oncology and their relative toxicity and to realize a quantitative model for the prediction of the number of new cases of the four tumors with the highest incidences (prostate, breast, lung and colon-rectal) in order to estimate the future drug expenditure in the United States. A characteristic element of this work is the study of complex models that are becoming increasingly important in the economic and health fields. The relationship between the toxicity and the cost of drugs was evaluated through the use of the cluster analysis and the Voronoi tessellation. We have also adopted an algorithm for the creation of an artificial neural network (ANN) that would allow, through the time-based learning of cancer-related risk factors, to estimate the incidence of the aforementioned malignancies up to 2050. The cost for a single patient was estimated considering an ideal patient with a height of 1.60 m and a weight of 60 kg eligible for receiving all the therapies currently available. Our software package Matlab R2014b. The cost/toxicity analysis showed the absence of a relationship between the increase of the cost of cancer drugs and the decrease in the rate of severe adverse events (SAE) and discontinuations (D) as evidenced by the presence of expensive drugs in the cluster 5 characterized by maximum toxicity. Analysis by ANN highlighted a decrease in the incidence of the four diseases, the largest of which in lung tumors (from 69 cases/100,000 inhabitants in 1992 to 32/100,000 in 2050). As for breast cancer, the estimated expenditure for the treatment of the two main histological types (HER2 positive and HER2 negative) will be in 2050 $ 1,435,532,808 and $ 4,421,355,552, respectively. Regarding immunotherapy, the expenditure related to the use of the three main agents will vary in 2050 from $ 1,616,529,467 (Pembrolizumab) to $ 2,782,155,725 (Nivolumab). In conclusion, our results focus on the need to optimize the evaluation of the cost/benefit ratio and cost/toxicity and underline how the amount of expenditure for the use of new generation drugs deserves careful evaluation in order to ensure their future sustainability.