Introduction: Breakthrough cancer pain (BTcP)A large proportion of patients with cancer suffer from Breakthrough cancer pain (BTcP). Several unmet clinical needs concerning BTcP treatment, like optimal opioids dosage, are being investigated. In this analysis the hypothesis whether distinct subtypes of BTcP exist and whether they can provide new insights into clinical practice is explored with an unsupervised learning algorithm. Methods: It was used partitioning around medoids algorithm on a large dataset of patients with BTcP previously collected by the Italian Oncologic Pain Survey (IOPS) group in order to identify possible subgroups of BTcP; the input of the algorithm consisted of different BTcP features, like its duration or its intensity. Silhouette statistics was used to pick an optimal number of clusters. Resulting clusters were analyzed in terms of BTcP therapy satisfaction, clinical features and usage of basal pain and rapid onset opioids. Opioids dosages were converted to a unique scale and BTcP-opioids-to-basal-pain-opioids ratio (OpR) was calculated for each patient. Polynomial logistic regression was used to catch non-linear relationships between therapy satisfaction and opioids usage. Results: The cohort comprised 4016 patients with controlled basal pain and suffering from BTcP. Our algorithm identified 12 distinct BTcP clusters. Optimal OpRs differed across the clusters, ranging from 15% to 50%. In the whole cohort, OpR was more clearly associated with therapy satisfaction compared with BTcP opioids or basal pain opioids alone. The majority of the clusters were linked to peculiar association of certain drugs with therapy satisfaction or dissatisfaction. A free online tool was created for new patients cluster computation (https://mancapaolo.shinyapps.io/UCBM_BTcPclusters/) in order to validate these clusters in future studies and to provide a possible, handy indications for personalized BTcP therapy. Discussion: This work proposes a classification for BTcP and identifies subgroups of patients with unique efficacy of different pain medications. This work supports the theory that the optimal dose of BTcP opioids depends on the dose of basal opioids and identifies novel values, possibly useful for future trials. These results will allow to target BTcP therapy based on patient characteristics and to define a “precision medicine” strategy also for supportive care.

A machine learning approach to identify clusters of patients with different Breakthrough cancer Pain (BTcP) clinical features and specific opioids response / Grazia Armento , 2020 Jul 09. 32. ciclo

A machine learning approach to identify clusters of patients with different Breakthrough cancer Pain (BTcP) clinical features and specific opioids response

2020-07-09

Abstract

Introduction: Breakthrough cancer pain (BTcP)A large proportion of patients with cancer suffer from Breakthrough cancer pain (BTcP). Several unmet clinical needs concerning BTcP treatment, like optimal opioids dosage, are being investigated. In this analysis the hypothesis whether distinct subtypes of BTcP exist and whether they can provide new insights into clinical practice is explored with an unsupervised learning algorithm. Methods: It was used partitioning around medoids algorithm on a large dataset of patients with BTcP previously collected by the Italian Oncologic Pain Survey (IOPS) group in order to identify possible subgroups of BTcP; the input of the algorithm consisted of different BTcP features, like its duration or its intensity. Silhouette statistics was used to pick an optimal number of clusters. Resulting clusters were analyzed in terms of BTcP therapy satisfaction, clinical features and usage of basal pain and rapid onset opioids. Opioids dosages were converted to a unique scale and BTcP-opioids-to-basal-pain-opioids ratio (OpR) was calculated for each patient. Polynomial logistic regression was used to catch non-linear relationships between therapy satisfaction and opioids usage. Results: The cohort comprised 4016 patients with controlled basal pain and suffering from BTcP. Our algorithm identified 12 distinct BTcP clusters. Optimal OpRs differed across the clusters, ranging from 15% to 50%. In the whole cohort, OpR was more clearly associated with therapy satisfaction compared with BTcP opioids or basal pain opioids alone. The majority of the clusters were linked to peculiar association of certain drugs with therapy satisfaction or dissatisfaction. A free online tool was created for new patients cluster computation (https://mancapaolo.shinyapps.io/UCBM_BTcPclusters/) in order to validate these clusters in future studies and to provide a possible, handy indications for personalized BTcP therapy. Discussion: This work proposes a classification for BTcP and identifies subgroups of patients with unique efficacy of different pain medications. This work supports the theory that the optimal dose of BTcP opioids depends on the dose of basal opioids and identifies novel values, possibly useful for future trials. These results will allow to target BTcP therapy based on patient characteristics and to define a “precision medicine” strategy also for supportive care.
9-lug-2020
Cancer pain; Machine learning; Breakthrough cancer pain (BTcP)
A machine learning approach to identify clusters of patients with different Breakthrough cancer Pain (BTcP) clinical features and specific opioids response / Grazia Armento , 2020 Jul 09. 32. ciclo
File in questo prodotto:
File Dimensione Formato  
DT_257_ArmentoGrazia.pdf

accesso aperto

Tipologia: Tesi di dottorato
Licenza: Creative commons
Dimensione 4.22 MB
Formato Adobe PDF
4.22 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12610/68808
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact