“…Overall, the methodological quality of included studies was moderate. There was moderate risk of bias for the domain of prognostic factor measurement in six studies, because five studies (Gallamini et al, 2007;Cerci et al, 2010;Straus et al, 2011;Markova et al, 2012;Zinzani et al, 2012) used a stand-alone PET system rather than integrated PET/CT, and one study did not report at all which PET system was used (Furth et al, 2011). In addition, there was moderate risk of bias for the domain of outcome measurement in five studies, because four studies (Gallamini et al, 2007;Markova et al, 2012;Filippi et al, 2013;Rossi et al, 2014) did not report the reference criteria for the confirmation of treatment failure and one study reported that histology was not used to confirm treatment failure in all cases (Straus et al, 2011).…”