By Lynda Williams, Senior medwireNews Reporter
Trial results for patients with steroid-refractory chronic graft versus host (SR-cGVHD) have overestimated the efficacies of treatments, suggests a meta-analysis published in The Lancet Haematology.
However, the authors believe that future trials could reduce this bias by following a 61-point checklist of guidelines, based on the US National Institute of Health (NIH) consensus project on SR-cGVHD in 2006.
The team identified 82 studies published between 1998 and 2013, assessing nine different treatments for SR-cGVHD, including extracorporeal photopheresis, rituximab, mycophenolate mofetil, imatinib, mesenchymal stem cells, methotrexate, pentostatin, sirolimus and thalidomide.
Meta-analysis of proportions for all studies reporting an overall response rate (ORR) gave an effect size of 66% with moderate heterogeneity, and the effect size for individual treatments varied from 50% for thalidomide to 79% for sirolimus.
Noting that the general methodological quality of the included studies was low, the researchers argue that “the pooled response rate shown in our meta-analysis should be interpreted as the theoretical average response rate perceived by readers of SR-cGVHD literature, rather than the true effect.”
They emphasise that the pooled ORR is “fairly high” compared with the low efficacy of treatments for SR-cGVHD previously reported in a European survey of clinicians treating haematopoietic stem cell patients, and in the knowledge that cGVHD mortality has not reduced in the past 3 decades.
“Consequently, we argue that the reported response for SR-cGVHD treatment has been overestimated”, say Attilio Olivieri, from Università Politecnica Delle Marche in Ancona, Italy, and co-authors.
The ORR did not significantly correlate with subsets of items used to define SR-cGVHD, or the specification of primary intervention or concomitant therapies, but was significantly associated with use of a subset of criteria for correct assessment of patient response.
Although trials published after 2006 were not significantly more likely to show global adherence to the NIH recommendations than earlier trials, the later studies were more likely to use the subset of criteria to correctly assess patient response and this, in turn, was associated with lower response rates.
“Better implementation of NIH recommendations might reduce false expectations about new interventions, and thus prevent clinical studies with ineffective treatments”, the team concludes.
Licensed from medwireNews with permission from Springer Healthcare Ltd. ©Springer Healthcare Ltd. All rights reserved. Neither of these parties endorse or recommend any commercial products, services, or equipment.