Does folate supplementation make sense in patients with rheumatoid arthritis treated with methotrexate?

The potential value ofthe antifolate methotrexate (N1O-methylaminopterin) in the treatment of rheumatoid arthritis (RA) was first recognised in 1951.' The low dose drug administration in divided doses 12 hours apart was developed for the treatment of patients with psoriasis, based on properties of psoriatic epithelial cells and the biological action of methotrexate on these cells.2 To date its mode of action has not been well established.3 It is generally believed that low dose weekly pulses of methotrexate are not markedly immunosuppressive. This is based on the inconsistency of in vitro and in vivo observed changes in immune variables.3 ' During treatment with low dose methotrexate, however, complications suggestive of immunosuppression may occur, such as varicella zoster, Pneumocstts carinni, and fungal infections.5 Only with a high dose of methotrexate, comparable with that used in patients with malignancies, does a decrease in surface DR expression on peripheralT lymphocytes occur.6 As activated T lymphocytes-bearing Ia antigens-have a role in the pathogenesis of rheumatoid arthritis7 this is an interesting finding. Biochemically the following mode of action of methotrexate is suggested: methotrexate competes with reduced folates for active transport into the cell and binds tightly (reversible) to dihydrofolate reductase, thus decreasing the intracellular formation of reduced folates with a consequent disturbance of thymidine, purine, and protein synthesis. Although the cellular basis for the action of methotrexate in RA remains unknown, it is presumed to be through its action on folate transport and metabolism. The hypothesis that a common folate dependent pathway is involved is reinforced by the fact that two antifolate drugs, methotrexate and sulphasalazine, are effective in the treatment of RA. Depletion of folate and inhibition of folate dependent enzymes may contribute to efficacy by altering inflammation and proliferation in arthritis target tissues,' but may also contribute to adverse drug effects of methotrexate in nontarget tissues. This is illustrated by the fact that methotrexate toxicity mimics clinical manifestations of folate deficiency, such as gastrointestinal intolerance, diarrhoea, stomatitis, and cytopenia. Unfortunately, serum folate concentrations merely reflect dietary intake rather than intracellular folate levels. Measurement of intracellular folate stores is indirectly possible by a biological assay of the activity of an enzyme system which synthesises serine from glycine and formate, and requires reduced folate coenzymes, the so called C1 index.8 As bone marrow and gastrointestinal tract have a high cell turnover, depending on a high intracellular supply of reduced folates needed for DNA, RNA, and protein synthesis, these non-target tissues may be the most sensitive to the adverse effects of methotrexate. The mechanism of methotrexate liver toxicity is not completely understood. Hepatocytes retain methotrexate in the active polyglutamated form for a prolonged period of time, which may augment toxicity.9 Mild adverse experience with low dose methotrexate is common in patients with RA, but continued treatment without methotrexate dose reduction is usually possible; in several five year follow up studies 50-60% of the patients continued to receive methotrexate with sustained clinical benefit.'1'2 As methotrexate is an effective second line drug for RA it is worthwhile questioning whether oral folate supplementation in rheumatoid patients treated with low dose methotrexate will decrease methotrexate toxicity without affecting its efficacy. Five studies'137 and at least two case reports'8 9 have been published dealing with folate supplementation in patients with RA treated with methotrexate; three studies dealt with folinic acid and two with folic acid supplementation (table). These prospective studies differ not only in study design (uncontrolled or double blind, placebo controlled), but also in study duration, relative dose of folic or folinic acid, and interval between supplementation and weekly pulse methotrexate. Except for Stewart's study the number of studied patients was low (seven to 32) and the duration of the study short (four to 48 weeks). Unfortunately, the Cl index was not measured in any of these studies. Hanrahan and Buckley's double blind, crossover study comparing folinic acid (leucovorin) with placebo did not show any decrease in toxic reactions in patients receiving folinic acid.'3 Is Tishler's unblinded case controlled study'4 showed folinic acid to be effective in preventing nausea caused by low dose methotrexate treatment in RA. This benefit was overshadowed, however, by increased rheumatic disease activity in all patients. Delaying folinic acid treatment for four to six hours after methotrexate administration should reduce competition for methotrexate uptake for entry into the cell. The extent of elimination of the uptake of methotrexate by folinic acid will be dependent on the relative dose of both drugs. The higher folinic acid dose in relation to methotrexate used by Tishler compared with the doses used by Hanrahan and Department of Rheumatology, Medical Centre of Leeuwarden, Henri Dunantweg 2, 8934 AD Leeuwarden, The Netherlands A A M E Stenger P M Houtman G A W Bruyn