It is now accepted that pivotal regulatory checkpoints of the immune response confer clinical therapeutic opportunities. Reductionist approaches to the identification of such nodes have conventionally relied upon murine disease models. Moreover, recent novel molecular and cellular approaches have facilitated the evaluation of human tissues from individuals with immune-mediated inflammatory diseases (IMIDs). Further progress will also be made using ex vivo cellular systems in which specific pathways can be modulated, as well as artificial intelligence and mathematical approaches.

However, to define the crucial function of a given moiety in the context of human IMIDs, there is no substitute for a clinical trial with a specific immune-targeted intervention. The seminal paper in this field, published in 1994, described the clinical benefits of a monoclonal antibody to tumour necrosis factor (TNF) in individuals with rheumatoid arthritis (RA). The preclinical rationale for this study was proposed by Ravinder Maini and Marc Feldmann on the basis of the identification of TNF and its properties in synovial tissues from patients with RA and successful targeting of TNF in mice. The implications of this trial were remarkable and supported the notion that functional cytokine hierarchies could be harnessed for clinical benefit. The convention of broad functional redundancy amongst cytokines was discarded, which had profound implications for future drug development.

In retrospect, several other cytokines, particularly IL-1, had similar levels of preclinical evidence to support their potential role as therapeutic targets in RA. However, IL-1 blockade did not have sufficient efficacy in the treatment of RA, which further highlights the crucial informative value of the TNF inhibition trial. Intriguingly, IL-1 blockade was subsequently found to have significant benefit in inflammatory monogenic disorders, other juvenile inflammatory syndromes and gout. Therefore, it seems that there are rarely ‘ineffective drugs’ but rather that cytokine inhibitors are effective only in the correct pathological context.

Deriving the preclinical rationale for progression to clinical trials remains a challenge even in 2020. Nevertheless, this first TNF inhibition study provided sufficient confidence in preclinical studies to drive a remarkable expansion of the field, resulting in biological therapeutics against several cytokines and their receptors, which variously confer clinical benefit across a range of IMIDs. Moreover, on the basis of such success, we can now seek a molecular taxonomy for IMIDs, based on clinical studies using specific cytokine inhibitors to define dominant immune pathways for each disease.