L Chim1, G Salkeld1, M R Stockler2,3,4, L Mileshkin5. 1. Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia. 2. Sydney Medical School, University of Sydney, Sydney, New South Wales, Australia. 3. National Health and Medical Research Council (NHMRC) Clinical Trials Centre, University of Sydney, Sydney, New South Wales, Australia. 4. Sydney Cancer Centre, RPA and Concord Hospital, Sydney, New South Wales, Australia. 5. Division of Haematology and Medical Oncology, Peter MacCallum Cancer Centre, Melbourne, Victoria, Australia.
Abstract
BACKGROUND: Little is known about the relative importance that oncologists attribute to the benefits and harms of anti-cancer drugs when considering treatment options with their patients. AIM: To quantify the trade-offs made between overall survival, progression-free survival and adverse effects. METHODS: A web-based survey elicited importance weights for the benefits and harms of bevacizumab or everolimus. Combining the importance weights with trial-based probabilities produced a score and ranking for each treatment option. RESULTS: A total of 40 responses was received for the bevacizumab scenario and 32 for the everolimus scenario. All respondents regarded overall survival and progression-free survival as the most important attributes - more important than avoiding the potential harms regardless of drugs. Among the potential harms, respondents allocated the highest mean importance weight to gastrointestinal (GI) perforation and rated absolute improvement in overall survival as 1.6 times and 2.3 times as important as avoiding GI perforation in the two versions of the bevacizumab scenario respectively. For the everolimus scenario, stomatitis and pneumonitis were allocated the highest mean importance weights with absolute improvement in overall survival rated as 2.2 times as important as avoiding stomatitis/pneumonitis. All 40 respondents (100%) favoured treatment option with bevacizumab to no bevacizumab based on respondents' determined weights for treatment attributes. The converse was found for everolimus with 22 (69%) of respondents preferring the 'no everolimus' option. CONCLUSION: Oncologists' preferences over the benefits and harms of treatment do, when combined with evidence of effect, influence treatment decisions for anti-cancer drugs.
BACKGROUND: Little is known about the relative importance that oncologists attribute to the benefits and harms of anti-cancer drugs when considering treatment options with their patients. AIM: To quantify the trade-offs made between overall survival, progression-free survival and adverse effects. METHODS: A web-based survey elicited importance weights for the benefits and harms of bevacizumab or everolimus. Combining the importance weights with trial-based probabilities produced a score and ranking for each treatment option. RESULTS: A total of 40 responses was received for the bevacizumab scenario and 32 for the everolimus scenario. All respondents regarded overall survival and progression-free survival as the most important attributes - more important than avoiding the potential harms regardless of drugs. Among the potential harms, respondents allocated the highest mean importance weight to gastrointestinal (GI) perforation and rated absolute improvement in overall survival as 1.6 times and 2.3 times as important as avoiding GI perforation in the two versions of the bevacizumab scenario respectively. For the everolimus scenario, stomatitis and pneumonitis were allocated the highest mean importance weights with absolute improvement in overall survival rated as 2.2 times as important as avoiding stomatitis/pneumonitis. All 40 respondents (100%) favoured treatment option with bevacizumab to no bevacizumab based on respondents' determined weights for treatment attributes. The converse was found for everolimus with 22 (69%) of respondents preferring the 'no everolimus' option. CONCLUSION: Oncologists' preferences over the benefits and harms of treatment do, when combined with evidence of effect, influence treatment decisions for anti-cancer drugs.
Authors: Henk Broekhuizen; Maarten J IJzerman; A Brett Hauber; Catharina G M Groothuis-Oudshoorn Journal: Pharmacoeconomics Date: 2017-03 Impact factor: 4.981