Search

Background: A clinical decision support system (CDSS) has been designed to predict the outcome (overall survival) by extracting and analyzing information from routine clinical activity as a complement to clinical guidelines in lung cancer patients. Materials and methods: Prospective multicenter data from 543 consecutive (2013–2017) lung cancer patients with 1167 variables were used for development of the CDSS. Data Mining analyses were based on the XGBoost and Generalized Linear Models algorithms. The predictions from guidelines and the CDSS proposed were compared. Results: Overall, the highest (> 0.90) areas under the receiver-operating characteristics curve AUCs for predicting survival were obtained for small cell lung cancer patients. The AUCs for predicting survival using basic items included in the guidelines were mostly below 0.70 while those obtained using the CDSS were mostly above 0.70. The vast majority of comparisons between the guideline and CDSS AUCs were statistically significant (p < 0.05). For instance, using the guidelines, the AUC for predicting survival was 0.60 while the predictive power of the CDSS enhanced the AUC up to 0.84 (p = 0.0009). In terms of histology, there was only a statistically significant difference when comparing the AUCs of small cell lung cancer patients (0.96) and all lung cancer patients with longer (≥ 18 months) follow up (0.80; p < 0.001). Conclusions: The CDSS successfully showed potential for enhancing prediction of survival. The CDSS could assist physicians in formulating evidence-based management advice in patients with lung cancer, guiding an individualized discussion according to prognosis. Instituto de Salud Carlos III PI16/02104 Junta de Andalucía PIN-0476-2017 Ministerio de Economía y Competitividad FPAP13-1E-2429

Resumen del trabajo presentado en el Meeting of the European Society for Radiotherapy and Oncology (ESTRO 2020), celebrado online del 28 de noviembre al 1 de diciembre, Purpose or Objective Prefoldin (PFDN) is a co-chaperone that contributes to both cytoplasmic and nuclear biological processes. Canonical PFDN has a heterohexameric jellyfish-like structure. Four ß-type subunits (PFDN1, 2, 4 and 6) form two dimers onto two subunits of the α type (PFDN3 and 5). PFDN2 and 6 are also components of the URI-prefoldin-like complex, which has been described to promote cancer. It has been shown that PFDN1 overexpresion promotes epithelial-mesenchymal transition (EMT) and lung cancer (LC) progression in different LC cell lines and murine models whereas cyclin A knockdown alone induces EMT and increases cell migration and invasion ability. We investigated whether this putative involvement of canonical PFDN in LC translates into the clinic. Material and Methods 58 non-small cell LC patients with available tumor tissue samples (59% squamous and 41% adenocarcinoma) were assessed. The stages were as follows: 24% I, 7% II, 61% III, and 8% IV. 90% of patients were primarily treated with surgery and 69% received chemotherapy. 86% underwent thoracic radiotherapy either primarily (41%) or after locorregional recurrence (45%). The levels of PFDN1, 3, 5 were examined by immunoblotting. Additionally, the mRNA expression of 518 LC cases from The Cancer Genome Atlas (TCGA) database was evaluated. To assess the risk of mortality and recurrences we used Kaplan-Meier and Cox proportional hazards analyses, Results PFDN1, 3, 5 and cyclin A overexpression (+++) were found in 22 (38%), 31 (53%), 24 (41%), and 14 (24%) tumor samples. After a follow up of 40 months, 39 (67%) patients were alive and 34 (58%) had experienced a recurrence (24 were distant metastasis). Body surface area and stage associated with overall survival (OS; p=0.01 and p=0.036, respectively), disease-free survival (DFS; p=0.033 and p=0.038, respectively), and distant metastasis-free survival (DMFS; p=0.002 and p=0.025, respectively) in the univariate analysis. In addition, the use of radiotherapy and chemotherapy also associated with DMFS (p=0.005 and p=0.015, respectively). PFDN1, 3 and 5 overexpression were associated with lower OS (p=0.002, p=0.015, and p=0.002, respectively), lower DFS (p=0.01, p=0.042, and p=0.055, respectively), and lower DMFS (p=0.011, p=0.036, and p=0.11, respectively). There was not any association with local recurrence. In the multivariate analysis, the PFDN5 retained significance for OS (HR 5.09; p=0.007) and the PFDN1 for DFS (HR 5.15; p=0.01) and DMFS (HR 5.45; p=0.05). In the TCGA adenocarcinoma cohort, there was a high correlation between PFDN1 and 5 (Pearson coefficient: 0.53; p

The development of healthcare patient digital twins in combination with machine learning technologies helps doctors in therapeutic prescription and in minimally invasive intervention procedures. The confidentiality of medical records or limited data availability in many health domains are drawbacks that can be overcome with the generation of synthetic data conformed to real data. The use of generative adversarial networks (GAN) for the generation of synthetic data of lung cancer patients has been previously introduced as a tool to solve this problem in the form of anonymized synthetic patients. However, generated synthetic data are mainly validated from the machine learning domain (loss functions) or expert domain (oncologists). In this paper, we propose statistical decision making as a validation tool: Is the model good enough to be used? Does the model pass rigorous hypothesis testing criteria? We show for the case at hand how loss functions and hypothesis validation are not always well aligned., This research was partly funded by the Spanish Ministry of Science, Innovation and Universities (AEI/FEDER, UE) grant numbers PGC2018-102145-B-C21 and PGC2018-102145-B-C22 (research project EDITH).

The digital twin in health care is the dynamic digital representation of the patient’s anatomy and physiology through computational models which are continuously updated from clinical data. Furthermore, used in combination with machine learning technologies, it should help doctors in therapeutic path and in minimally invasive intervention procedures. Confidentiality of medical records is a very delicate issue, therefore some anonymization process is mandatory in order to maintain patients privacy. Moreover, data availability is very limited in some health domains like lung cancer treatment. Hence, generation of synthetic data conformed to real data would solve this issue. In this paper, the use of generative adversarial networks (GAN) for the generation of synthetic data of lung cancer patients is introduced as a tool to solve this problem in the form of anonymized synthetic patients. Generated synthetic patients are validated using both statistical methods, as well as by oncologists using the indirect mortality rate obtained for patients in different stages. Ministerio de Ciencia, Innovación y Universidades PGC2018-102145-B-C21 Ministerio de Ciencia, Innovación y Universidades PGC2018-102145-B-C22 Unión Europea 825619 (AI4EU)