This 1998 single-centre observational study at Royal Melbourne Hospital compared antibody responses to influenza vaccination in three groups: 20 healthy controls, 25 renal transplant recipients on prednisolone/cyclosporine/azathioprine (PCA), and 13 recipients on prednisolone/cyclosporine/MMF (PCM), using blinded haemagglutination assays on a single vaccine batch .
The effect is largest for A/H1N1 (post-vaccination PCM vs PCA p=0.0007; vs control p<0.0001) and is corroborated by a second clinical endpoint: 76% of PCM patients (10/13) failed a β₯4-fold rise to any antigen, versus 36% (9/25) for PCA and 25% (5/20) for controls .
Strengths: Blinded serology, a single vaccine batch, three complementary endpoints (fold-rise, protective-titre conversion, GMT), baseline equivalence between transplant groups, and internal consistency with mechanistic plausibility (MMF blocks de novo purine synthesis, relatively lymphocyte-specific) .
Blindspots and confounders: Non-randomized, unmatched group allocation (PCM patients came from the Tricontinental trial; PCA largely did not); age differed across groups (48.0 vs 35.6 years), and age independently reduces vaccine response β the authors assert this is unlikely to matter without formal adjustment. The PCM group (n=13) is small, yielding low power, which the authors themselves concede for the influenza B analysis. Pre-vaccination titres were lower in immunosuppressed groups (e.g., A/H1N1 β₯40 in 40% control vs 7.6% PCM), creating ceiling/floor effects. Drug exposure timing and trough MPA levels were not measured, so doseβresponse inference is absent .
Interpretation caution: The design cannot isolate MMF from possible allocation differences; the authors' proposed application to antibody-mediated disease (SLE, IgA nephropathy) rests on two cited case reports plus mechanistic inference β conjecture, not demonstrated causal benefit. The claim that MMF patients are not at greater infection risk "thus far" relies on trials not powered for that endpoint, an omission-bias risk. Downstream literature indirectly supports broader utility: MMF later proved superior to cyclophosphamide for lupus nephritis remission (RR 1.72, 95% CI 1.17β2.55) .
The central finding β MMF profoundly suppresses the vaccine antibody response in humans in vivo β is well-supported for influenza A antigens and is mechanistically coherent, but the modest, non-randomized sample and unmeasured confounders (age, MPA exposure) temper causal certainty; a randomized, MPA-monitored vaccine-response trial would be the decisive test.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.