AFFILIATIONS
1 The Hospital for Sick Children, Toronto, Ontario, Canada
2 University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, United States
3 Ambroise-Paré Hospital, Assistance Publique–Hôpitaux de Paris (AP-HP), Boulogne-Billancourt, France, Versailles Saint-Quentin-en-Yvelines University (UVSQ), Versailles, France
4 Hospital Nacional de Pediatria Garrahan, Buenos Aires, Argentina
5 Mayo Clinic, Rochester, Minnesota, United States
6 The Hospital for Sick Children, Toronto, Ontario, Canada
7 The Hospital for Sick Children, Toronto, Canada
8 The Hospital for Sick Children, Toronto, Ontario, Canada
9 The Hospital for Sick Children, Toronto, Ontario, Canada
10 The Hospital for Sick Children, Toronto, Ontario, Canada
11 Hospital Universitario Cruces, Barakaldo, Spain
12 Rigshospitalet Copenhagen University Hospital, Copenhagen, Denmark
13 Hospital Nacional de Pediatria Garrahan, Buenos Aires, Argentina
14 Dalhousie University, Halifax, Canada
15 London Health Sciences Centre, London, Canada
16 Rostock University Medical Centre, Rostock, Germany
17 Dana-Farber Cancer Institute and Boston Children's Hospital, Boston, Massachusetts, United States
18 Rigshospitalet Copenhagen University Hospital, Copenhagen, Denmark
19 Hospital Nacional de Pediatria Garrahan, Buenos Aires, Argentina
20 University of Alabama at Birmingham, Birmingham, Alabama, United States
21 Krankenhaus der Elisabethinen, Graz, Austria
22 CHU Sainte-Justine, Quebec, Canada
23 University Hospital and Faculty of Medicine, Masaryk University, Brno, Czech Republic
24 Dalhousie University, Halifax, Canada
25 Ain Shams University Hospital (ASU), Cairo, Egypt
26 Showa Inan General Hospital, Komagane, Japan
27 The Hospital for Sick Children, Toronto, Ontario, Canada
28 Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands
29 Cerrahpasa Faculty of Medicine, Istanbul University–Cerrahpasa, Istanbul, Turkey
30 Meyer Children’s Hospital IRCCS, Florence, Italy
31 Valley Children's Hospital, Madera, California, United States
32 King Abdulaziz Medical City, Riyadh, Saudi Arabia
33 Shenzhen Children's Hospital, Shenzhen, China
34 Meyer Children’s Hospital IRCCS, Florence, Italy
35 Pathology Children’s Hospital of Pittsburgh of UPMC, Pittsburgh, Pennsylvania, United States
36 The University of Alabama at Birmingham, Birmingham, Alabama, United States
37 Motol and Homolka University Hospital, Prague, Czech Republic
38 Yedei Chesed, New York, New York, United States
39 Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands
40 Beijing Children’s Hospital,, Beijing, China
41 The Hospital for Sick Children, Toronto, Ontario, Canada
42 The University of Alabama at Birmingham, Birmingham, Alabama, United States
43 The Hospital for Sick Children, Toronto, Ontario, Canada
44 Memorial Sloan Kettering Cancer Center, New York, New York, United States
ABSTRACT
Background: Rosai-Dorfman-Destombes disease (RDD) is a rare myeloid neoplasm with diverse clinical features and prognosis. It is characterized pathologically by the accumulation of CD68/S100-positive and CD1a-negative histiocytes with frequent emperipolesis. Recent molecular studies showed that KRAS and MAP2K1 mutations were present in one-third of RDD patients. Due to its rarity, the best treatment strategy is not well defined.
Methods: The IRHDR is an international prospective and retrospective registry for rare histiocytic neoplasms, including RDD. We enrolled 267 confirmed RDD patients diagnosed between 1995-2026. Demographic, pathological/molecular, clinical, treatment and outcome data were collected.
Results: Median age at diagnosis was 34 years (range 0-80; IQR 45.25), 151 (57%) were females, 105 were children (<18 years, 39%) and 162 adults (>18 years, 61%). One-hundred-four (39%) had multisystemic (MS), 94 (35%) extra-nodal single-site (SS) and 69 (26%) nodal only RDD. Lymph nodes were most commonly involved (45% overall; 60% in children and 40% in adults), followed by bone (29%), skin (28%), CNS (14%), ENT (10%), ocular (7%), gastrointestinal (GI, 5%), genitourinary (GU, 4%), respiratory (4%), liver (3%), and cardiovascular (3%). Among 174 tested patients, 88 (51%) tested positive for molecular mutations; KRAS was most frequent (n=30), followed by MAP2K1 (n=26), BRAF (n=9; 7 were V600E), NRAS (n=4), SLC29A3 (n=3), MAP2K2 (n=3), TET2 (n=3), and ARAF (n=2). Forty-three patients (16%) had associated/pre-existing malignancy, 17 (6%) associated autoimmune disease, and 3 (1%) H syndrome. Initial management included surgery (30%; 77% were SS and 23% MS-RDD), steroids (26%), observation (17%), chemotherapy (13%), MEK inhibitors (MEKi, 13%), rituximab regimens (5%), mTOR inhibitors (1%), IL-6 inhibitors (1%) and radiotherapy (1%). Higher overall response rates (ORR) were seen with MEKi (88%), followed by rituximab (77%), surgery (75%), mTOR inhibitors (75%), chemotherapy (69%), steroids (61%), and radiotherapy (50%). Among 210 initially treated patients, 88 (42%) relapsed or were refractory to initial therapy. Seventy patients received steroids as first-line therapy either alone (40 patients; 55% relapsed) or with another therapy (30 patients; 50% relapsed). Eighty-one patients had surgery as first-line therapy alone (65 patients; 34% relapsed) or with other therapies (16 patients; 31% relapsed). Among 26 MAP2K1+ cases, 6 received first-line MEKi with 100% ORR. Among all relapsed patients, higher ORR were seen with steroids (70%), MAPKi (68%) and cladribine/clofarabine (62%). Three RDD patients were salvaged with lenalidomide/dexamethasone with 100% ORR. Among 43 initially observed patients (17 nodal, 18 SS, 8 MS), only 3 (7%) required second-line therapy at median follow-up of 14 months. At median follow-up of 28 months (range, 0.5-358), 8 (3%) patients died: 4 from progressive disease (2 CNS, 1 MS, 1 nodal), 3 RDD-associated malignancies and 1 COVID-19 infection. Five-year overall survival (OS) was 95%. Five-year progression-free survival (PFS) was 85%; univariate analysis showed liver (HR 5.34, p = 0.028, 95% CI 1.20-23.70), MAP2K1 mutation (HR 3.04, p = 0.037, 95% CI 1.07-8.67), and bone (HR 2.50, p = 0.049, 95% CI 1.00-6.24) as statistically significant predictors of relapse; however, GU site (HR 3.41, p = 0.054, 95% CI 0.98-11.90) and receipt of second-line therapy (HR 3.03, p = 0.051, 95% CI 1.00-9.19) showed higher risk, while KRAS mutation yielded a lower risk of relapse (HR 0.62, p = 1.00, 95% CI 0.14-2.67). In a multivariate Cox model, both MAP2K1 mutation (HR 3.84, p = 0.021, 95% CI 1.19-10.13) and GU site (HR 4.16, p = 0.029, 95% CI 1.50-15.08) remained independent predictors of relapse.
Conclusions: This study is the largest contemporary and molecularly characterized series of RDD patients to date. RDD presents with heterogeneous clinical and molecular phenotypes. Best initial treatment response rates are seen with MEK and mTOR inhibitors, and rituximab. Although mortality from RDD is low, more than a third of the patients had relapsed/refractory disease following initial therapy, while recurrence rate after observation strategy was low. MAPKi, cladribine/clofarabine and lenalidomide/dexamethasone are effective as salvage therapies. Liver and bone sites and MAP2K1 mutation were associated with relapse. These data can potentially form the basis of future treatment recommendations for RDD patients.