GOYA

Obinutuzumab is a glycoengineered type II anti-CD20 antibody that had already beaten rituximab on progression-free survival in chronic lymphocytic leukaemia and in follicular lymphoma (GALLIUM). GOYA tested the same substitution in untreated advanced diffuse large B-cell lymphoma, randomising 1,418 patients to eight 21-day cycles of obinutuzumab (706) or rituximab (712) with six or eight cycles of CHOP.

After a median observation of 29 months the investigator-assessed progression-free survival events were almost identical, 201 (28.5 per cent) with obinutuzumab and 215 (30.2 per cent) with rituximab, a stratified hazard ratio of 0.92 (95 per cent confidence interval 0.76 to 1.11, p = 0.39), with three-year rates of 70 and 67 per cent. Independently reviewed progression-free survival, the other time-to-event endpoints and response rates were all similar. Grade 3 to 5 adverse events were more frequent with obinutuzumab (73.7 against 64.7 per cent), as were serious adverse events (42.6 against 37.6 per cent) and fatal events (5.8 against 4.3 per cent).

GOYA matters because it broke the assumption that an antibody improvement carries across B-cell lymphomas. The exploratory finding that germinal-centre B-cell tumours did better than activated B-cell tumours irrespective of treatment is the clearest signal the trial left behind, and it points at the genetic subtype work rather than at the antibody. The registry lists the study as terminated, which reflects the end of a programme that did not continue rather than a safety stop.