The number of gadolinium (Gd)-enhancing lesions on monthly magnetic resonance imaging (MRI) scans of the brain is widely used to monitor multiple sclerosis (MS) clinical trials. In relapsing-remitting (RR) MS, serial monthly Gd-enhanced MRI scans are five to ten times more sensitive than clinical measures in detecting disease activity and, therefore, allow treatment effects to be investigated by studying fewer patients for shorter follow-up periods than when using clinically-based end points. In addition, counting the numbers of enhancing lesions is a very reproducible process in experienced hands, and it also allows the observers to be unaware of the treatment regimen of individual patients, thus avoiding bias due to unblinding. A retrospective study by Auer et al recently reported that the frequency and extent of Gd-enhancing lesions on serial MRI scans from MS patients is influenced by seasonal fluctuations, being significantly higher in the first than in the second half of the year. This study was, however, based on relatively small samples of patients and scans (202 scans were obtained from 53 patients, ie, an average number of less than four scans per patient) and the frequency of scanning was highly variable. Therefore, its results might have been heavily influenced by MRI findings from individual patients. Because the MRI activity at a given timepoint is significantly correlated with that seen during the previous and the subsequent months, a cluster of few scans with very high numbers of Gd-enhancing lesions could have artificially increased the average activity during certain periods. The month-to-month fluctuations of the number of enhancing lesions might, therefore, be a chance effect depending on the variable timing of patient sampling and on the high interpatient variability of MRI activity. Because Auer et al concluded that such a marked seasonal variation of Gd-enhancing lesions in MS patients may prevent us from reaching reliable conclusions from MRImonitored trials, we re-addressed this issue by studying the seasonal fluctuations of the number of Gd-enhancing lesions as seen on 11 consecutive brain MRI scans obtained every 4 weeks over a 10-month period from 120 RRMS patients, who were part of the untreated arm of a previous multi-center, randomized, double-blind, placebo-controlled trial. All the scans were collected between February 1997 and August 1998, using a standardized imaging protocol across the participating centers. The only treatment allowed during the study period was steroid for acute relapses. Enhancing lesions were counted by two experienced observers by consensus. The average number of enhancing lesions per scan was 4.1 (SD 5 7.1). The Figure shows the month-to-month variation of the mean number of enhancing lesions, which is higher in March [mean number of enhancing lesions (SD) 5 5.7 (9.5)] and lower in December [mean number of enhancing lesions (SD) 5 3.3 (5.0)]. To avoid interpretation bias due to the interpatient variability of MRI activity, we made a statistical analysis of the pairwise comparisons between the numbers of enhancing lesions during the four seasons in 92 patients who had at least one scan for each of the seasons (Friedman test for nonparametric data). The mean numbers of enhancing lesions (SD) were 4.4 (6.6), 5.1 (8.5), 4.5 (6.7), and 4.5 (7.6) for winter, spring, summer, and autumn, respectively. No significant difference of MRI activity between seasons was found. These results demonstrate that, although MRI activity in RRMS patients varies in the different seasons, this fluctuation is not significant. Our data were obtained from a large sample of patients with heterogeneous geographical origins, who were recruited from different European and Canadian centers. In addition, baseline MRI scans were collected over a period covering all the four seasons. For these reasons, our results should not be biased by other factors that are known to influence MS activity. In conclusion, the seasonal fluctuations of subclinical activity in patients with RRMS should not affect the interpretation of the results from MRImonitored clinical trials a great deal.