Diagnostic Yield and Risks Associated With Retroperitoneal Biopsy in Erdheim‐Chester Disease

Erdheim-Chester disease (ECD) is a rare adult-onset histiocytosis, often characterized by multi-organ infiltration with histiocytes, frequently driven by activating MAPK pathway mutations [1, 2]. Diagnosis requires tissue biopsy for histological and molecular analysis, but histology is often nonspecific, and confirmation depends on mutation detection. Identifying the best biopsy target is crucial, balancing diagnostic yield and procedural risk. Cutaneous lesions, present in only 20%–30% of patients, are preferred but rare [3, 4]. Bone lesions are almost universal (95%), but retroperitoneal involvement (65%) is a frequent and accessible target. Some affected organs (CNS, heart, and aorta) are unsuitable for biopsy [4, 5]. We aimed to evaluate the diagnostic performance, safety, and predictive factors of retroperitoneal biopsy in ECD. We retrospectively reviewed all adults referred to Pitié-Salpêtrière Hospital (Paris, France) between May 2015 and March 2024 for suspected ECD with retroperitoneal involvement. Diagnosis was established according to international guidelines [1]. Clinical parameters (age at diagnosis, sex, and detailed organ involvement), biological parameters, and pathological and molecular analyses of biopsy specimens were obtained from medical records. Clinical, biological, histological, and molecular data were extracted from medical records. All patients provided written informed consent. Biopsies were performed during short-term hospitalization, after confirmation of hemostatic compatibility (platelets > 50 × 109/L; prothrombin rate > 70%; aPTT < 1.4; discontinuation of anticoagulation/dual antiplatelet therapy). Procedures were conducted under CT guidance by senior interventional radiologists, with patients in prone position under local anesthesia (1% xylocaine). A coaxial technique with core biopsy needle (automatic or semi-automatic) was used, and ≥ 2 samples were collected (Figure 1 and Figure S1 [Tangential perirenal biopsy technique]). One sample was fixed in formalin, the other kept fresh for molecular testing. Patients were monitored for ≥ 3 h post-procedure and observed overnight. Adverse events were graded using Common Terminology Criteria for Adverse Event Classification (CTCAE) (5th version, 2017). Technical success required ≥ 2 samples of ≥ 1 cm. Biopsy yield was defined as contributive if it enabled histological or molecular diagnosis. Clinical and imaging predictors were analyzed, including perirenal infiltration thickness (Table S1) circumferential involvement, and FDG-PET uptake (SUVmax). Associations with yield were assessed by χ2, Student's t, Mann–Whitney U tests, and logistic regression. p < 0.05 was considered significant. From May 2015 to April 2024, 80 patients underwent retroperitoneal biopsy: 67 men (83.8%) and 13 women (16.2%), mean age 69.5 ± 12.5 years, BMI 27.5 ± 4.5 kg/m2. Only one patient was on specific ECD treatment before biopsy. Sixty-six patients (82.5%) had a single biopsy, 14 (17.5%) had two, and two (2.5%) had three. Automatic devices were used in 60 cases (75%), semi-automatic in 20 (25%). Needle gauges were 16 G (57%), 18 G (38%), and 14 G (5%). Seventy-six biopsies (95%) were perirenal and four (5%) peri-aortic. Technical success was 100%. Among first biopsies, 60 (75%) were contributive and led to histological confirmation of ECD. Molecular analysis was performed in 55 (91.7%) of these, identifying BRAFV600E in 35 (58.3%) and MAP2K1 in 7 (11.7%). Second biopsies were performed in 14 patients, mainly for diagnostic confirmation (n = 11) or mutation testing (n = 3). They were contributive in 10 (71.4%). Of two third biopsies, one was contributive. Overall, ECD was confirmed in 69 patients. Genetic analysis was possible in all: BRAFV600E mutations in 37 (53.6%) and MAP2K1 in 9 (14.5%) (Table 1). Four patients (5%) experienced complications: three grade 1 perirenal hematomas requiring no intervention, and one grade 3 anemia after peri-aortic biopsy, necessitating transfusion. No major long-term complication occurred. Mean perirenal infiltration thickness was 10.9 ± 5.6 mm; infiltration was circumferential in 64 patients (84.2%). Distribution of infiltration grades was: grade IV (42.1%), grade III (39.5%), grade II (14.5%), and grade I (3.9%). FDG-PET was positive (SUVmax ≥ 3.0) in 45 patients (59.2%), with mean uptake 4.6 ± 2.6. Biopsies targeted PET-avid areas in 39 (86.7%) of these patients. Univariate analysis showed contributiveness was associated with circumferential perirenal infiltration (OR = 11.0, p = 0.01), thickness > 10 mm (OR = 5.6, p = 0.09), and biopsy in PET-avid zone (OR = 6.9, p = 0.009). In multivariate analysis, circumferential infiltration (OR = 5.79, p = 0.019) and PET-avid targeting (OR = 5.01, p = 0.05) remained independent predictors (Table 2). Neither age, sex, BMI, biopsy device, needle gauge, nor number of samples influenced yield. No factor predicted complications. This study is the largest to date assessing retroperitoneal biopsies in suspected ECD. It demonstrates a high diagnostic yield (90% overall), with sufficient tissue for genetic testing in all confirmed cases. MAPK pathway mutations were detected in 70% of patients, consistent with previous series [1, 2]. Diagnostic yield was strongly influenced by target selection. Circumferential perirenal infiltration and PET-avid zones were independent predictors of contributiveness. These findings align with oncology studies linking FDG-PET positivity to biopsy success [6]. However, some patients had positive biopsies despite PET negativity, showing that PET should not exclude biopsy when retroperitoneal targets are present. Infiltration thickness < 10 mm and focal involvement were associated with lower yield, suggesting that alternative biopsy sites should be considered in such cases. Technical parameters, including device type, needle gauge, and number of samples, did not significantly impact diagnostic performance. Two samples appeared sufficient for histology and molecular testing. Target selection thus outweighs technical variations in determining biopsy success. Association between perirenal infiltration and biopsy positivity may also be explained by the commonly employed “tangential” approach, which facilitates the retrieval of an adequate tissue sample while minimizing complications (especially renal injuries). Limitations include the retrospective design, single-center setting, and modest sample size, inherent to the rarity of ECD. Referral bias to a tertiary center may also explain the high prevalence of confirmed cases. Retroperitoneal biopsies, particularly perirenal, provide high diagnostic yield and low complication rates in suspected ECD. Circumferential infiltration and PET-guided targeting improve success, while technical parameters are less influential. In the absence of cutaneous lesions, the retroperitoneum should be considered the preferred biopsy site for both histological confirmation and molecular characterization. Study conception and design: A.R., M.P., C.R., and J.H. Acquisition of data and performance of experiments: A.R., M.P., J.-F.E., F.C., J.L.R., P.-L.C., R.D., S.B., A.R., P.-A.V., F.C.-A., Z.A., C.R., and J.H. Analysis of data, drafting and writing of the manuscript: A.R., M.P., C.R., and J.H. All the authors were involved in reviewing the manuscript and approved the final version for publication. The authors have nothing to report. This research was performed without patient involvement. Patients were not invited to comment on the study design and were not consulted for the development of patient-relevant outcomes or for interpretation of the results. Patients were not invited to contribute to the writing or editing of this document for readability or accuracy. The authors have nothing to report. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Table S1: Categorical classification of perirenal infiltration on computed tomography scan in patients with Erdheim-Chester disease. Figure S1: Erdheim-Chester disease: tangential perirenal fat biopsy technique. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.