Clinical Manifestations and Treatment Response of Patients With Syndrome of Undifferentiated Recurrent Fever (SURF)

Systemic autoinflammatory diseases (AIDs) are a heterogeneous group of inborn errors of immunity characterized by episodes of sterile inflammation without evidence of pathogenic autoantibodies or autoreactive T lymphocytes [1]. The most known manifestation of AIDs is recurrent fever, often associated with inflammatory involvement of joints, brain, eyes, skin, and serous membranes. Peripheral serum inflammatory markers, as C reactive protein and serum amyloid A, always increase during attacks [2]. AIDs are rare (incidence of < 1/2000 individuals) [3], although they are increasingly recognized worldwide. Most AIDs are caused by a single genetic variant, although some may result from somatic mosaicism or low-penetrance variants [4, 5]. Genetic testing is necessary for molecular diagnosis, which is confirmed in patients with typical symptoms and pathogenic variants [1].

Among patients with clinical signs of AIDs, at least 40% have no molecular diagnosis, and approximately 50% have no pathogenic variants in AID-related genes analyzed with targeted next-generation sequencing panels [6]. Most of these patients meet classification criteria for periodic fever, aphthous stomatitis, pharyngitis, and adenitis (PFAPA) syndrome [7]. PFAPA syndrome is the most common polygenic AID, characterized by recurrent episodes of oral ulcers, bilateral cervical lymphadenopathy, and/or exudative tonsillitis. PFAPA syndrome has a self-limiting clinical course, with remission during late childhood. In cases of frequent and/or severe inflammatory attacks that cause significant discomfort to the patient, tonsillectomy is suggested, with a high rate of efficacy (75%–90%). Diagnostic criteria for PFAPA syndrome have been published elsewhere [8].

Another group of patients presenting with recurrent fever and negative or inconclusive genetic tests for AIDs has been classified under the term syndrome of undifferentiated recurrent fever (SURF) [9]. SURF patients do not carry variants in monogenic AIDs-associated genes and do not fulfill the diagnostic criteria for PFAPA syndrome or other inherited recurrent fever syndromes. Despite evolving knowledge of the molecular mechanisms underlying AIDs, it remains unclear whether SURF represents a distinct clinical entity with complex, polygenic inheritance or a heterogeneous group of patients with similar manifestations and what is the most effective treatment for these patients. In this study, we performed a systematic review with meta-analysis to highlight the clinical features and the best treatment options in SURF patients.

SURF is an umbrella clinical entity recently included in AIDs and defined by recurrent fever attacks without any overt infectious disease, absent criteria for PFAPA syndrome, and negative genetic tests for monogenic AIDs. This clinical and laboratory classification of exclusion allows clinicians to study and follow a group of patients who present evidence of autoinflammation without treatment guidelines. The term allows to clearly differentiate these patients from those with evidence of tonsils as triggers of inflammatory flares even in the case of incomplete PFAPA criteria, also known as incomplete PFAPA patients, for which tonsillectomy is commonly effective [32].

Key findings of our study are reported in Table 1. The present study highlights the efficacy rate of long-term use of colchicine and anakinra in previously published SURF cohorts. Further studies in different ethnic groups are warranted to confirm this observation. Furthermore, a regular follow-up aimed at capturing evidence of newly emerging symptoms suggestive of a well-defined AID and a regular genetic re-screening are warranted. In fact, the number of monogenic diseases associated with recurrent fever is increasing, and some SURF patients can be molecularly defined with new sequencing techniques (i.e., multiplex ligation-dependent probe amplification, long-read sequencing) in the future. In the meantime, SURF classification allows for the study more deeply, even with omics techniques, a condition that appears to be polygenic in nature.

Our study has several limitations. Some authors included in SURF cohorts also included patients without overt fever during flares (about 10%), suggesting different manifestations used to suspect an autoinflammatory process (i.e., positive inflammatory markers, etc.). These criteria of suspicion are not usually reported, introducing a possible enrolment bias affecting the interpretation of pooled data. Furthermore, different techniques of genetic sequencing used across different studies may also impact the SURF classification. We hope future studies can follow the proposed empirical indications for the clinical suspicion of SURF in order to limit this bias [33].

A standardized definition of treatment response is lacking, and the data used in our analysis may reflect different definitions between studies. Notably, information regarding dose and duration of treatments are often incomplete, influencing the pooled estimates. These factors should be taken into account when interpreting our results. We think the EUROFEVER registry criteria should be used to define treatment response in AID as well as in SURF patients, standardizing future studies [34].

E.L. and R.P. designed the study and analyzed the data; F.B. performed the statistical analysis; E.L. wrote the first draft of the manuscript; all authors reviewed and approved the final version as submitted.

The authors have nothing to report.

The authors declare no conflicts of interest.