This cohort study assesses outcomes after treatment with cobimetinib based on MAPK pathway alterations in patients with Rosai-Dorfman disease.
Key Points
Question
What are the outcomes with cobimetinib, a MEK inhibitor, treatment in KRAS- or MEK-variant Rosai-Dorfman disease (RDD)?
Findings
In this cohort study of 16 patients with RDD who were treated with cobimetinib, somatic alterations in the KRAS or MEK genes were detected in 50% of patients. Patients with KRAS/MEK alterations had significantly higher overall response rate (88% vs 38%), deeper responses (71% vs 0% complete responses), and better progression-free survival (at 1 year, 100% vs 29% were free from progression or death, respectively) compared with those without such alterations.
Meaning
Cobimetinib is associated with positive outcomes in KRAS- or MEK-variant RDD.
Abstract
Importance
Rosai-Dorfman disease (RDD) is a rare histiocytic neoplasm with recent studies showing alterations in the MAPK pathway, most commonly in the KRAS and MEK genes in about 40% of patients. Reports on the use of MEK-inhibitor therapy in RDD have been limited to small case studies. There are no approved treatments for this neoplasm, and therefore patients with RDD need efficacious treatments.
Objective
To study the outcomes after treatment with cobimetinib based on MAPK pathway alterations in patients with RDD.
Design, Setting, and Participants
This retrospective cohort study conducted at 2 tertiary care centers included patients with RDD who underwent treatment with cobimetinib between January 1, 2013, and December 1, 2021. Cobimetinib was administered at a dosage of 20 to 60 mg orally once daily as a single agent for 21 days in a 28-day cycle. Pathology was centrally reviewed. Response assessment was centrally conducted and was based on the established positron emission radiography response criteria used for clinical trials of targeted therapies in histiocytosis.
Main Outcomes and Measures
Main outcomes were overall response rate (ORR), progression-free survival (PFS), adverse events (AEs) of cobimetinib in the entire cohort, and ORRs and PFS based on MAPK pathway alterations in patients with RDD.
Results
A total of 16 patients (median [range] age at cobimetinib initiation, 57 [31-74] years; 11 [69%] women) were included in the study. The median follow-up duration was 19.0 months (95% CI, 8.4-27.8 months). The ORR was 63% (n = 10), including 5 complete responses and 5 partial responses. Somatic alterations in the KRAS or MEK genes were detected in 8 (50%) patients. Patients with KRAS or MEK alterations had significantly higher ORR (88% vs 38%; P = .03), deeper responses (complete responses among responders: 71% vs 0%; P = .002), and better PFS (at 1 year, 100% vs 29% were free from progression or death, respectively; P < .001) compared with those without such alterations. Grade 2 or higher AEs occurred in 12 (75%) patients, and 9 (56%) required dose reduction or temporary/permanent treatment discontinuation due to AEs.
Conclusions and Relevance
In this cohort study, treatment with cobimetinib was associated with positive outcomes in KRAS- or MEK-variant RDD. However, AEs requiring dose modifications were common.
Introduction
Rosai-Dorfman disease (RDD) is a rare histiocytic neoplasm, which histopathologically presents with infiltration of large, pale histiocytes bearing characteristic immunophenotype (CD68±/CD163+/S100+/CD1a−/langerin−) and frequent emperipolesis in a lymphoplasmacytic background.1,2 Although the classic description of RDD is massive cervical lymphadenopathy, extranodal RDD involving any organ system has been reported to occur more commonly in contemporary studies.1,3 Cutaneous, retroperitoneal, central nervous system (CNS), intrathoracic, gastrointestinal, and head and neck involvement are the commonly involved extranodal sites. Subcutaneous masses, palpable lymphadenopathy, lytic bone lesions, and sinus involvement are the most common clinical presentations.3 Not all patients with RDD require treatment, as some are asymptomatic with an indolent course, and a minority of patients will have spontaneous resolution. For those who need treatment, surgical or radiation therapy can be used for localized diseases. For systemic disease, chemotherapy and immunomodulators have been used with variable efficacy.1
Recent studies have identified somatic alterations in the mitogen-activated protein kinase (MAPK) pathway, especially in Kirsten-rat-sarcoma-viral-oncogene homolog (KRAS) and MAPK (MAP2K1 or MEK) genes in 35% to 40% of patients with RDD.4,5,6,7 However, reports on the use of MEK-inhibitor therapy in RDD are very limited. Cobimetinib, a MEK inhibitor, produced an overall response rate (ORR) of 89% in a phase 2 study, which predominantly enrolled patients with Erdheim-Chester disease.8 This trial included only 2 patients with RDD. Therefore, it is necessary to determine the outcomes of MEK inhibition in a larger cohort of patients with RDD. To address this, we present the clinical outcomes of a larger series of patients with RDD treated with cobimetinib outside of a clinical trial setting.
Methods
Patients with RDD who were consecutively seen at Mayo Clinic and University of Alabama at Birmingham from January 2013 to December 2021 and treated with cobimetinib single-agent treatment were included. The study was approved by the respective institutional review boards, and all data were deidentified; hence, patient informed consent was waived. Pathology was centrally reviewed (K.L.R. and D.M.M.). Response assessment was centrally conducted (J.R.Y.) by using the established positron emission radiography response criteria used in histiocytosis (eMethods in the Supplement).8,9 Among patients without positron emission radiography–avid lesions, consensus radiographic response criteria for histiocytosis were used.9 Cobimetinib dosing ranged from 20 to 60 mg orally once daily for 21 days in a 28-day cycle. Adverse events (AEs) were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0. Genomic analysis was conducted using target-capture next-generation sequencing (NGS) through Tempus-xT assay or StrataNGS assay platforms. See eMethods in the Supplement for additional details.
Results
Of the 89 patients with RDD seen during the study period, 16 (18%) were treated with cobimetinib and were included. Cobimetinib was administered frontline in 7 patients. The median (range) age at cobimetinib initiation was 57 (31-74) years, and most (11 [69%]) were women. Among patients who received cobimetinib in the subsequent line, the median (range) number of prior treatments was 3 (2-10) (eTable 1 in the Supplement). All patients had extranodal disease. The clinical and genomic characteristics are shown in Table 1. The median follow-up was 19.0 months (95% CI, 8.4-27.8 months), and the median (range) duration of cobimetinib treatment was 5.5 (2.0-23.1) months. The ORR for the entire cohort was 63% (n = 10), and 5 patients achieved a complete response (CR) and a partial response (PR), respectively (Figure, A; eFigure in the Supplement). The 6- and 12-month progression-free survival (PFS) rates were 78% and 65%, respectively (Table 2). The response rates were similar when cobimetinib was given in the first or a subsequent line (71% vs 56%; P = .51).
Table 1. Baseline Characteristics and Treatment Outcomes of Cobimetinib.
| Patient No./age, ya | Dose, mg | Organ(s) involved | Alterations | Response | Duration of treatment, mo | Status | Secondary cancers |
|---|---|---|---|---|---|---|---|
| 1/40s | 60 | CNS | None | PD | 2 | Died | None |
| 2/70s | 20 | Bone, lymph node, sinus | KRAS p.T58I, CDKN2A p.16INK4a A20S | CR | 19+ | Alive | None |
| 3/70s | 20 | Kidney, RP skin, testis | MAP2K1 p.F53L, ASXL1 p.T822fs, CDK11B copy number loss | SD | 6 | Alive | None |
| 4/40s | 20 | Bone, kidney, lymph node, RP | PIK3CA p.N345K | SD | 7 | Alive | None |
| 5/60s | 60 | Skin, trachea | KRAS p.K117N | CR | 23+ | Alive | None |
| 6/40s | 60 | Bone | None | SD | 4 | Alive | None |
| 7/50s | 40 | Bone, CNS, eye, heart, kidney | MAP2K4 p.S16R | CR | 17+ | Alive | None |
| TP53 p.F134C | |||||||
| CREBBP p.D639fs | |||||||
| 8/60s | 20 | Bone, eye, skin | SMARCA4 p.K953fs, BRG1 | PD | 6 | Died | MDSc |
| 9/30s | 20 | Bone, lymph node | None | PR | 3 | Alive | None |
| 10/70s | 20 | Bone, CNS, kidney | KRAS p.A146T | CR | 5 | Alive | None |
| 11/60s | 20 | Kidney, RP | None | PR | 6+ | Alive | None |
| 12/30s | 40 | Skin, subcutaneous tissue | None | SD | 10 | Alive | None |
| 13/50s | 40 | Bone, lymph node, skin | MAP2K1 p.F53L | PR | 2+ | Alive | None |
| 14/40sb | 40 | Breast, pelvic soft tissue | KRAS p.A146P | PR | 4 | Alive | None |
| 15/40s | 40 | Bone, pelvic soft tissue | None | PR | 3+ | Alive | None |
| 16/50s | 60 | Bone, kidney | MAP2K1 p.D67N | CR | 9+ | Alive | None |
Abbreviations: CNS, central nervous system; CR, complete response; MDS, myelodysplastic syndrome; PD, progressive disease; PR, partial response; RP, retroperitoneum; SD, stable disease.
Age decade at the time of cobimetinib initiation.
Response assessment based on clinical evaluation because the patient had claustrophobia and a repeat positron emission radiography scan could not be performed for response assessment.
Diagnosed 6 years following the diagnosis of Rosai-Dorfman disease.
Figure. Disease Course and Outcomes in the Study Cohort.
A, Swimmer plot depicting disease course, treatment outcomes, and adverse events (AEs) in patients treated with cobimetinib. B, Progression-free survival in patients with and without KRAS/MEK alterations.
Table 2. Comparative Outcomes of Cobimetinib Based on KRAS/MEK Alterations.
| Variable | Median (95% CI) | P value | ||
|---|---|---|---|---|
| Entire cohort | Patients without KRAS/MEK alterations (n = 8) | Patients with KRAS/MEK alterations (n = 8) | ||
| Follow-up (95% CI), mo | 19.0 (8.4-27.8) | 25.0 (2.8-55.0) | 14.0 (2.4-23.1) | .13 |
| Response, No. (%) | 10 (63) | 3 (38) | 7 (88)a | .03 |
| Progression-free survival, mo | 21.4 (5.1-24.8) | 12.0 (2.0-21.4) | 24.8 (NR-NR) | .008 |
| 6 mo; 12 mo, % | 78; 65 | 58; 29 | 100; 100 | NA |
| Duration of response, mo | NR (4.0-NR) | 3.0 (NR-NR) | NR (NR-NR) | .01 |
| Event-free-survival, mo | 6.4 (4.0-10.0) | 5.6 (2.1-10.0) | 7.9 (4.0-NR) | .20 |
| Overall survival, mo | NR (13.2-NR) | 30.0 (13.2-NR) | NR (NR-NR) | .28 |
| 6 mo; 12 mo, % | 100; 90 | 100; 83 | 100; 100 | NA |
| Time to best response, median (range), mo | 3.6 (2.4-5.6) | 3.3 (1.0-NR) | 3.7 (1.5-5.6) | .64 |
Abbreviations: NA, not applicable; NR, not reached.
One patient had stable disease and did not progress on therapy.
Twelve (75%) patients were started at a dose lower than 60 mg at the discretion of the treating physician. Responses were seen among the 7 patients who started at 20 mg (2 CRs, 2 PRs) and 5 patients who started at 40 mg (1 CR, 3 PRs). The ORRs were similar between patients who started at lower doses compared with full dose (67% vs 50%; P = .55). Grade 2 or higher toxic effects appeared to be less frequent among those who started at lower doses (67% vs 100%; P = .09).
Alterations in KRAS and MEK genes were detected in 8 (50%) patients (Table 1 and Figure, A). The ORR was significantly higher among patients with KRAS/MEK-variant RDD compared with patients with no such alterations (88% vs 38%; P = .03). Among responders, the CR rate was 71% in the KRAS/MEK-variant group vs 0% in those without KRAS/MEK alterations (P = .002; Figure, A). The median PFS for KRAS/MEK-variant group was significantly longer than for those without such alterations and at 1 year, 100% vs 29% were free from progression or death, respectively (P < .001; Table 2 and Figure, B).
Treatment-related AEs led to dose reductions, treatment withholding, and/or permanent discontinuation in 9 (56%) patients (Figure, A; eTable 2 in the Supplement). Grade 2 or higher toxic effects were similar among patients with and without KRAS/MEK alterations (75% vs 75%; P > .99). Cobimetinib was permanently discontinued in 9 (56%) patients—5 due to AEs and 4 due to disease progression (Figure, A). At last follow-up, 14 (88%) patients were alive (Figure, A; Table 1). One patient died of therapy-related myelodysplastic syndrome. The patient had received multiple cytotoxic treatments before initiation of cobimetinib. Another died secondary to progression of RDD in the CNS.
Discussion
This is, to our knowledge, the largest reported series of patients with RDD treated with cobimetinib. Cobimetinib had positive outcomes in the entire cohort, with the majority of the responses seen among cases with KRAS/MEK alterations. Objective responses were observed even with low doses of cobimetinib, which caused fewer AEs. There are only 8 published RDD cases treated with cobimetinib, with variable outcomes.8,10,11,12,13 Of these patients, only 3 had assessment of MAPK pathway alterations. One patient with KRASG12R alteration achieved a PR. The other 2 patients did not have KRAS/MEK alterations; 1 attained PR while the other had stable disease.8,10
Interestingly, in our study, close to a third of patients without detectable MAPK pathway alterations had a response to cobimetinib, albeit only PRs. A possible reason for this finding could be false-negative NGS results due to sparse pathologic histiocytes in biopsy specimens. Another explanation could be that inhibition of the MAPK pathway through cobimetinib may be associated with decreased monocyte- and macrophage-induced inflammation, an active pathobiological component in RDD.14 Due to the rarity of RDD, the reported outcomes of chemotherapy, steroids, surgery, or immunomodulators are mostly limited to case series, hindering accurate assessment of treatment-related end points. Overall, corticosteroids have an average response rate of close to 50%. Radiation therapy has been used to treat localized disease with response seen in about 25% of patients.1,3
Limitations
Our study is limited by the relatively small sample size and its retrospective nature, and AEs may be underestimated. We were not able to capture quality-of-life data during cobimetinib treatment or provide patient-reported outcomes.
Conclusions
This cohort study demonstrates that cobimetinib treatment was associated with positive outcomes in patients with KRAS/MEK-variant RDD. Similar to the prospective study of cobimetinib in histiocytic neoplasms, dose-limiting toxic effects were common irrespective of alteration profile.8 Therefore, close monitoring of AEs while on therapy is essential. Patients who were treated with low doses of cobimetinib also had remarkable clinical responses. Further studies are needed to confirm activity of cobimetinib in patients with other bona fide MAPK pathway alterations and to evaluate alternate targeted therapies with better toxicity profile. Given that CNS-involved RDD may have detrimental outcomes, future studies should also focus on MEK inhibitors with improved CNS penetrance. Additionally, long-term follow-up will be necessary to establish the optimal duration of therapy with MEK inhibitors like cobimetinib.
eMethods.
eFigure. Radiographic Depiction of the Complete Metabolic Response in a Patient With KRAS Mutated Rosai-Dorfman Disease Treated With Cobimetinib
eTable 1. Sequence of Prior Treatments Before the Initiation of Cobimetinib
eTable 2. Adverse Effects of Cobimetinib
References
- 1.Abla O, Jacobsen E, Picarsic J, et al. Consensus recommendations for the diagnosis and clinical management of Rosai-Dorfman-Destombes disease. Blood. 2018;131(26):2877-2890. doi: 10.1182/blood-2018-03-839753 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703-1719. doi: 10.1038/s41375-022-01613-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Goyal G, Ravindran A, Young JR, et al. ; Mayo Clinic Histiocytosis Working Group . Clinicopathological features, treatment approaches, and outcomes in Rosai-Dorfman disease. Haematologica. 2020;105(2):348-357. doi: 10.3324/haematol.2019.219626 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Diamond EL, Durham BH, Haroche J, et al. Diverse and targetable kinase alterations drive histiocytic neoplasms. Cancer Discov. 2016;6(2):154-165. doi: 10.1158/2159-8290.CD-15-0913 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Lee LH, Gasilina A, Roychoudhury J, et al. Real-time genomic profiling of histiocytoses identifies early-kinase domain BRAF alterations while improving treatment outcomes. JCI Insight. 2017;2(3):e89473. doi: 10.1172/jci.insight.89473 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Garces S, Medeiros LJ, Patel KP, et al. Mutually exclusive recurrent KRAS and MAP2K1 mutations in Rosai-Dorfman disease. Mod Pathol. 2017;30(10):1367-1377. doi: 10.1038/modpathol.2017.55 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Matter MS, Bihl M, Juskevicius D, Tzankov A. Is Rosai-Dorfman disease a reactve (sic) process? detection of a MAP2K1 L115V mutation in a case of Rosai-Dorfman disease. Virchows Arch. 2017;471(4):545-547. doi: 10.1007/s00428-017-2173-4 [DOI] [PubMed] [Google Scholar]
- 8.Diamond EL, Durham BH, Ulaner GA, et al. Efficacy of MEK inhibition in patients with histiocytic neoplasms. Nature. 2019;567(7749):521-524. doi: 10.1038/s41586-019-1012-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Goyal G, Tazi A, Go RS, et al. International expert consensus recommendations for the diagnosis and treatment of Langerhans cell histiocytosis in adults. Blood. 2022;139(17):2601-2621. doi: 10.1182/blood.2021014343 [DOI] [PubMed] [Google Scholar]
- 10.Jacobsen E, Shanmugam V, Jagannathan J. Rosai-Dorfman disease with activating KRAS mutation—response to cobimetinib. N Engl J Med. 2017;377(24):2398-2399. doi: 10.1056/NEJMc1713676 [DOI] [PubMed] [Google Scholar]
- 11.Moyon Q, Boussouar S, Maksud P, et al. Lung involvement in Destombes-Rosai-Dorfman disease: clinical and radiological features and response to the MEK inhibitor cobimetinib. Chest. 2020;157(2):323-333. doi: 10.1016/j.chest.2019.09.036 [DOI] [PubMed] [Google Scholar]
- 12.Giuffrè C, Giuffrè G. Choroidal involvement in Rosai-Dorfman disease successfully treated with cobimetinib. Indian J Ophthalmol. 2020;68(9):2051-2053. doi: 10.4103/ijo.IJO_1359_20 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.López-Aldabe K, Escrihuela-Vidal F, Tuells-Morales M, et al. Multiple drug regimen-refractory Rosai-Dorfman-Destombes disease mimicking relapsing polychondritis successfully treated with cobimetinib. Eur J Case Rep Intern Med. 2022;9(2):003076. doi: 10.12890/2022_003076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kaminska B. MAPK signalling pathways as molecular targets for anti-inflammatory therapy—from molecular mechanisms to therapeutic benefits. Biochim Biophys Acta. 2005;1754(1-2):253-262. doi: 10.1016/j.bbapap.2005.08.017 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eMethods.
eFigure. Radiographic Depiction of the Complete Metabolic Response in a Patient With KRAS Mutated Rosai-Dorfman Disease Treated With Cobimetinib
eTable 1. Sequence of Prior Treatments Before the Initiation of Cobimetinib
eTable 2. Adverse Effects of Cobimetinib
