Cancer treatment usually induces a depression of the immune system, that may cause neutropenic periods in which the patient is exposed to an increased risk of infections and sever complications, typically announced by fever (common symptom). A prospective observational study conducted between 2002 and 2004 at the Gaslini Hospital in Italy gathered information on neutropenic periods of 157 children affected by different types of solid tumours, receiving chemotherapy regimens (classified into two broad categories, very intensive and intensive) and, possibly, autologous hemopoietic stem cell transplantation. The aim was to asses the role of treatment in the incidence of recurrent febrile complications during neutropenic periods, accounting for the history of past events, the intermittence of exposure and the time-varying nature of treatment. We considered a stratified Cox model for a marginal and frailty analysis, stratified by type of cancer, with age and sex as fixed covariates and treatment as a time-dependent covariate. The results suggested that patients who underwent transplantation were more likely to develop febrile episodes. Unmeasured heterogeneity between patients did not appear to be relevant and results of the frailty model were very similar to that of the marginal model. The need to evaluate how the history of previous febrile episodes impacts on the current risk of developing fever requires further developments. In particular, the approach explored in this regard is the Aalen’s Additive Hazard model, that may be useful to disentangle the role of treatment and the influence of previous events.

Analysis of recurrent complications with intermittent exposure and time-varying treatment in cancer / P. Rebora, F. Bagnasco, M.G. Valsecchi, R. Haupt. ((Intervento presentato al 29. convegno Annual Conference of the International Society for Clinical Biostatistics tenutosi a Copenhagen nel 2008.

Analysis of recurrent complications with intermittent exposure and time-varying treatment in cancer

P. Rebora
Primo
;
2008

Abstract

Cancer treatment usually induces a depression of the immune system, that may cause neutropenic periods in which the patient is exposed to an increased risk of infections and sever complications, typically announced by fever (common symptom). A prospective observational study conducted between 2002 and 2004 at the Gaslini Hospital in Italy gathered information on neutropenic periods of 157 children affected by different types of solid tumours, receiving chemotherapy regimens (classified into two broad categories, very intensive and intensive) and, possibly, autologous hemopoietic stem cell transplantation. The aim was to asses the role of treatment in the incidence of recurrent febrile complications during neutropenic periods, accounting for the history of past events, the intermittence of exposure and the time-varying nature of treatment. We considered a stratified Cox model for a marginal and frailty analysis, stratified by type of cancer, with age and sex as fixed covariates and treatment as a time-dependent covariate. The results suggested that patients who underwent transplantation were more likely to develop febrile episodes. Unmeasured heterogeneity between patients did not appear to be relevant and results of the frailty model were very similar to that of the marginal model. The need to evaluate how the history of previous febrile episodes impacts on the current risk of developing fever requires further developments. In particular, the approach explored in this regard is the Aalen’s Additive Hazard model, that may be useful to disentangle the role of treatment and the influence of previous events.
2008
International Society for Clinical Biostatistics (ISCB)
Analysis of recurrent complications with intermittent exposure and time-varying treatment in cancer / P. Rebora, F. Bagnasco, M.G. Valsecchi, R. Haupt. ((Intervento presentato al 29. convegno Annual Conference of the International Society for Clinical Biostatistics tenutosi a Copenhagen nel 2008.
Conference Object
File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate

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/2434/58722
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact