We retrospectively evaluated a cohort of patients with CAPS under long-term anti-IL-1 therapy with special emphasis on renal outcome. Patients in this study were predominantly female and tended to be older than in other studies (3,9–11). Most previous studies involving CAPS patients were carried out in a paediatric population. Thus, our cohort was very different with a longer exposition to the natural course until diagnosis of CAPS and subsequently a higher risk of developing sequelae. However, as in previous reports, musculoskeletal symptoms, rash and fever were the most common symptoms in the present study (3). As described before, inflammatory parameters and clinical symptom load improved markedly after start of anti-IL-1 therapy and remained suppressed throughout the follow-up period (10).
Our data show that CKD is a relevant complication of CAPS, affecting 4 out of 21 patients (19%) at various ages at time of diagnosis. In an earlier study by Kümmerle-Deschner et al., frequency of renal affection in CAPS was reported to be as high as 59% (12). In that study, however, most patients had a rather rare mutation (E311K) with very high frequency of renal involvement (77%) as opposed to patients having other mutations (47%). In addition, no patients with the R260W mutation were included, which is the predominant mutation in our study and presumably the most frequent CAPS mutation per se (3). Notably, the authors did not define renal involvement and KDIGO criteria were not used; essentially, cut-off values of GFR and proteinuria defining CKD were not reported.
All patients with significant CKD in the present study belonged to the same family bearing the R260W mutation. Surprisingly, only 50% of family members had CKD. This finding suggests not only a familial predisposition independent of the specific disease-causing mutation but also the presence of additional factors contributing to renal injury in CAPS patients. Albeit KDIGO stages were generally associated with age at diagnosis, the risk for amyloidosis does not appear to be a simple derivative thereof, nor simply the result of the inflammatory load: In fact, the youngest patient with CKD was most seriously affected requiring renal replacement therapy from age 24 on. On the contrary, the oldest patient in that family, aged 46 at time of initial diagnosis, did not show any signs of CKD. This broad inter-patient variability in developing amyloidosis is in line with earlier reports (13).
Anti-IL-1 therapy is the gold standard of therapy for CAPS (5). In our cohort, it resulted in profound and sustained reduction in inflammatory parameters as well as symptom load in all patients. During follow-up, however, CKD progressed in two patients despite anti-inflammatory and clinical efficacy of anti-IL-1 therapy, while patients without significant kidney disease at therapy start did not develop CKD stages G3-5 or A3. We concluded that neither age and serum parameters before therapy initiation nor clinical or serological efficacy during therapy may predict renal outcome once CKD is present.
Progress of CKD despite clinical and serological efficacy of anti-IL-1β therapy, in principle, leaves two explanatory options: First, amyloidosis might be at least in part due to a mechanism independent of IL-1 in the first place. This assumption is supported by low serum inflammatory parameters in patients with renal amyloidosis before therapy initiation but would contradict earlier findings of a landmark study on amyloidosis in autoinflammatory disease (4). Second, and more compelling, renal disease might be already in a state of self-perpetuating damage, irrespective of primary cause. In favour of the latter argument the fact remains that in those patients no other signs of aggravating amyloid deposition were apparent in an observational period of 8 and 13 years, respectively. Furthermore, the favourable outcome of the young transplanted patient with anti-IL-1β therapy points to efficacy in preventing rather than reversing manifest amyloidosis. This assumption is finally corroborated by the fact that under therapy no new cases of CKD developed in the observational period.
The present study has some limitations, foremostly small sample size. However, despite data being per se limited to few individuals in orphan diseases, it is the first time that renal involvement in CAPS has been systematically studied. Frequency of renal affection has to be taken cautiously, especially as only a proportion of 61% of patients had identified family members with CAPS. In those families all carriers of the respective genes had been identified. However, it is likely that in the remainder cases not all gene carriers had been identified as CAPS is pre-eminently an autosomal dominant disease. This might lead to overreporting if patients with only mild symptoms might not be admitted to our clinic. Another limiting factor lies in the “unknown known” of patients not suffering from significant kidney disease in our study and who have not rigorously been examined for the presence of amyloidosis, neither by biopsy nor by SAP-scintigraphy. Hence, we cannot definitely draw conclusions, whether in those patients, subclinical amyloid deposition, if any, might be reversed or progression halted by treatment.
In conclusion, we showed that in CAPS patients: (1) CKD is relatively common; (2) development of CKD may be due to a familial risk beyond the mutational status and/or the result of a potential second hit; (3) anti-IL-1 therapy significantly improves inflammatory parameters and symptom burden; (4) development of CAPS-associated CKD may be prevented by therapies targeting IL-1, while renal outcome may not be predicted once renal damage is already present. Overall, manifestation and progression of CKD in CAPS is possibly preventable by early therapy initiation, hence emphasizing recognition and treatment of the disease in a timely fashion.