The challenge posed by the question, ‘When to initiate and discontinue biological treatments for rheumatoid arthritis?’ , can best be met by the recognition that ‘there is a need for more pharmacogenetic studies to possibly predict response in patients who are candidates for biological therapy’ , and that the identification of the promoter region of the tumour necrosis factor gene  has created such an opportunity and provided the basis for the hypothesis that ‘cytokine (and/or receptor) gene SNPs (single-nucleotide polymorphisms) might, indeed, play a role in determining the biological effects, hence the clinical effectiveness of [biological] therapies’ . Using a genome-wide strategy, investigators in the United Kingdom have validated this hypothesis by identifying the association of seven genetic loci with the response to anti-tumour necrosis factor therapy in rheumatoid arthritis (RA) . A subgroup of patients arguably justifying more focussed pharmacogenetic profiling might be those patients with RA who already appear to belong to a phenotypically distinct subtype, characterised by male predominance and by high prevalence of pleural effusion , and pericarditis, respectively [7, 8]. In some of those patients, notably those with pleural effusion, genetic profiling has already resulted in the documentation of high prevalence of the HLA-B8 and HLADW3 genotypes . An additional feature which distinguishes patients with either pleural disease or pericardial disease from the rest of the RA population is the fact that pleural effusion  and pericarditis , respectively, may antedate the onset of arthritic symptoms, sometimes by as long as 2.5 years  or 2 years , respectively. What needs to be evaluated is whether or not there is altered responsiveness to biological treatments either in patients with RA who either have clinically overt pleural effusion or pericarditis, or in their RA counterparts who have subclinical pleural or pericardial disease, given the fact that there is mounting evidence that biological treatments which have succeeded in controlling arthritic symptoms, sometimes have the paradoxical effect of precipitating the emergence of clinically overt pleural [12, 13] or pericardial disease [14, 15], perhaps indicative of a ‘dichotomy’ between biological treatment effects on articular versus extra-articular manifestation of RA . This dichotomy was exemplified by a 71-year-old man in whom etanercept achieved complete remission of arthritis but, nevertheless, subsequently gave rise to the occurrence of bilateral pleural effusions, which resolved after discontinuation of this drug, and did not recur on 6 months follow-up . A corollary to this anecdote is the report of two patients, in addition to five cited in the literature, in all seven of whom treatment with biological agents was associated with new onset of clinically overt pericardial effusion, despite good control of joint symptoms . Lessons learnt from pharmacogenetic profiling of specific subgroups such as men with pleural effusions or pericardial effusions, which are either subclinical or clinically overt, might add an extra dimension to our understanding of the relationship between genotype and responsiveness to biological treatments in RA.
Conflict of interest statement
No conflict of interest to declar.