Assessment of clinical trial inclusion criteria to a cohort of lymphoma patients undergoing autologous transplantation: are trial results generalized to other patient populations?

The Parma trial (Philip, N Engl J Med, 1995) demonstrates superior progression-free (PFS) and overall (OS) survival in patients with chemo-sensitive relapsed aggressive histology lymphoma (AHL) who receive high-dose therapy and stem cell transplantation (SCT). We suspect this trial guides therapy in patients not meeting its strict eligibility criteria. The aim of our study was to assess how the Parma trial's results apply to patients referred to our centre for SCT, and whether outcomes differ between patient groups accord .ng to the potential eligibility for that trial. Between 09/1995 and 03/1999, 60 patients with refractory/ relapsed AHL were referred. Two expert reviewers blinded to treatment received and outcome, independently assessed cases to determine potential Parma eligibility, und whether in their judgement SCT should be offered. Disagreements were resolved by consensus; PFS and OS were compared in the determined groups. Of 60 patients referred. 27 (45%) received SCT. At2yrsPFS (32% vO%;p<.0001) and OS (41% v4%;p<. 0001) were superior in transplanted patients. Reviewers agreed on Parma eligibility criteria in 47'60 (78%) cases. Consensus was reached in 13 after clarification of pathology (5) and treatment response (8). Among those receiving SCT, 5/27 (19%) were Parma 'eligible'; 22/27 (81%) did not meet Parma inclusion criteria for reasons that included primary refractory disease (26%), failure to achieve a partial response (20%), salvage therapy other than DHAP (4f,%) and age60 (30%). A comparison of outcomes beteen Parma eligible and ineligible groups demonstrated no detectable difference in 2 yr PFS (p=0.38) or OS (p=0.41). Among those receiving SCT, 17/27 (63%) were judged appropriate for SCT; 10/27 (37%) were judged inappropriate. No differences were detected in 2 yr PFS (32% v 15%; p=0.3) or OS (48% v 30%; p=0.14) between groups judged appropriate or inappropriate to be offered SCT. We conclude that the results of the Parma trial are generalized to many patients who would not have been eligible for that trial. A consensus review process detects patients who receive SCT who do not meet standard criteria for offering this treatment. Further studies are requ jed to determine whether SCT benefits these patient groups.