Methotrexate and cytarabine for adult patients with newly diagnosed Langerhans cell histiocytosis: A single arm, single center, prospective phase 2 study

Abstract

Langerhans cell histiocytosis (LCH) is a rare, heterogeneous histiocytic disorder occurring in patients of all ages. The current standard treatment protocol for children with de novo multisystem LCH is vinblastine plus prednisone.1 This regimen has never been proven effective for adults in a prospective study. Moreover, our previous report showed the reactivation frequency was 71%-78.6% after vindesine and prednisone-based therapy.2 These data indicate the concept of treatment for children probably cannot be translated to adults. A retrospective study showed an advantage for cytarabine monotherapy compared with vinblastine/prednisone in bone LCH patients.3 But no consensus on the best first-line treatment strategy has yet been reached. Oral methotrexate had been successfully used in skin LCH previously.4 Considering the relatively high frequency of pituitary involvement and late onset of neurodegenerative symptoms, patients may benefit from the combination of cytarabine and methotrexate, as both these drugs cross the blood-brain barrier. The successful experience of combination methotrexate/cytarabine used in non-Hodgkinlymphoma afforded us a reference point for choosing the dosage of methotrexate.5 This phase 2, prospective, single-center study was designed to evaluate the efficacy and safety of a methotrexate and cytarabine (MA) treatment regimen in adults. This is with newly diagnosed multisystem disease (MS) LCH or LCH with multifocal single system (SS-m) involvement. We enrolled all patients 18 years of age or older with newly diagnosed MS or SS-m LCH at Peking Union Medical College Hospital (Beijing, China). The methotrexate/Ara-C (MA) regimen was administered every 35 days for six cycles in total. Methotrexate 1 g/m2 was administered by 24-hour infusion on day 1 and Ara-C 0.1g/m2 by 24-hour infusion for 5 days. Patients underwent clinical assessment, physical examination, laboratory tests and imaging as appropriate for the LCH location after three and six cycles of MA, and then every 3 months for disease evaluation. Imaging data were collected from 18F-fluorodeoxyglucose positron emission tomography (FDG-PET), thoracic and abdominal computed tomography (CT) and cerebral magnetic resonance imaging (MRI). Disease was assessed using standard evaluations, as defined by Histiocyte Society criteria.1 If all signs and symptoms were resolved, the patients were classified as having non-active disease (NAD). Otherwise, they were classified as having active disease (AD). Active disease was further subdivided into regressive disease (signs and symptoms were improved, AD/better), stable disease (SD, persistence of signs and symptoms) or progressive disease (PD, progression and/or appearance of new lesions). Toxicities were recorded and graded according to the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (CTCAE 4.0). The overall response rate (ORR) was defined as the proportion of patients who had a NAD or AD/better response to MA therapy. Both SD and PD were considered to be poor responses. The primary endpoint was event-free survival (EFS). Events were defined as a poor response to MA, reactivation after MA therapy and death from any cause. Event-free survival was defined as the duration from initiation of MA treatment to the date of a first event or last follow-up. Overall survival (OS) was defined as the duration from initiation of MA treatment to the date of death or last follow-up. The final follow-up date was 30 June 2019. The study is registered with ClinicalTrials.gov, number NCT 02389400. From January 2014 to December 2018, 83 patients were eligible for the study. The baseline characteristics of the 83 patients are summarized in Table S1. Overall, 49 patients were male (59.0%). The median age was 33 years (range, 18-65 years). At the start of the study, six patients were SS-m LCH (7.2%), 77 patients were MS LCH (92.8%). The median number of involved organs was three (range 1-8). The most common organ involved in the total cohort was bone (78.3%), followed by lung (67.5%), pituitary (62.7%), lymph nodes (38.6%), liver (27.7%), skin (22.9%), thyroid (19.3%), spleen (13.3%) and CNS (3.6%). No patients had hematopoietic system involvement. All patients received at least one course of MA therapy, with a median of six courses (range 1-6). Overall, 69 patients (83.1%) completed protocol treatment, and 14 patients (16.9%) went off protocol (13 patients' decision, one poor response). This is illustrated in the flow diagram (Figure 1A). The ORR was 87.9%, including 43 patients (51.8%) classified as having NAD and 30 patients (36.1%) as AD/better. Five patients (6.0%) were evaluated as SD and five as PD. After a median follow-up of 23 months (range 7-66 months), one patient died of disease progression (23 months after initiation of MA treatment), and 25 patients had reactivation of the disease. In 15 cases, the reactivation was limited to a single organ, including three cases in skin, one in pituitary, two in lung, five in liver and four in bone. Ten patients reactivated in multiple systems, including four with pituitary. Taken together five patients (6.0%) reactivated in the pituitary. No degenerative CNS disease was observed. The estimated 3-year OS and EFS were 97.7% and 68.0%, respectively (Figure 1B). To evaluate the prognostic factors of EFS using univariate analysis, patients with liver involvement at baseline had significantly shorter EFS compared to patients without liver involvement (15 months [95% CI 0-31.0] versus not reached, P <.001.) (Figure 1C). Patients with spleen involvement also had significantly shorter EFS compared to patients without spleen involvement (15 months [95% CI 3.6-26.4] versus not reached, P = .004). Lung involvement also indicated poor EFS (39 months [95% CI 21.1-56.9] versus not reached, P = .021), as did skin involvement (23.3 months [95% CI 6.5-40.1] versus not reached, P = .031. No statistically significant difference in EFS was observed based on bone, pituitary, lymph node or thyroid involvement at baseline. The EFS was also evaluated using a multivariate Cox regression model, and liver involvement remained predictive of poor EFS (P = .012; HR 0.339, 95% CI 0.146-0.784) (Table S2). The most common toxicity was hematological adverse events (Table S3). All patients experienced neutropenia and thrombocytopenia. Thirty-five patients (42.2%) had grade 4 neutropenia; 43 patients (51.8%) had grade 3 neutropenia. Fourteen patients (16.9%) had grade 4 thrombocytopenia; 13 patients (15.7%) had grade 3 thrombocytopenia. Platelet transfusions were given in 14 patients (16.9%). The median duration of grade 3 to 4 myelosuppression was 6 days (0-10 days). No patients received prophylactic antimicrobial treatment during any of the cycles. Forty patients (48.2%) experienced febrile neutropenia, including 38 (45.8%) grade 3 and 2 (2.4%) grade 4. The most common non-hematological toxicities were gastrointestinal complications. Forty-two patients developed nausea, and two of them were grade 3. Vomiting occurred in nine patients. All gastrointestinal complications improved with supportive treatment. Alanine aminotransferase increases occurred in 19 patients (22.9%), with grade 3 events in two patients. Fourteen patients (16.9%) experienced grade 1 urticaria. One patient had a grade 2 creatinine increase, and achieved a full recovery. One patient (1.2%) had SPM (oral squamous cell carcinoma) 56 months after the last course of the MA regimen. No treatment-related deaths occurred. The features of LCH are well described in children; however, they remain poorly defined in adults. There are no standard therapies for adult LCH. Treatment recommendations are based on retrospective studies and small case series. To our knowledge, this study is the first prospective phase 2 trial undertaken in adults with newly diagnosed LCH. In this prospective study, the MA regimen gave an ORR of 87.9% in 83 adult LCH patients as first-line treatment, which was higher than those of vinblastine and prednisone based-therapy, as well as cytarabine or cladribine monotherapy.3, 6 These results also revealed a significant improvement in EFS and OS rates in this cohort compared to historical results. The 32% probability of LCH recurrence at 3 years found in our study, is similar to the 34% 3-year probability of recurrence observed in the prolonged 12-month treatment arm, in risk organ-negative child patients in the LCH-III study.1 Note, MS LCH with risk organ involvement indicated worse prognosis. Organs at risk include the hematopoietic system, spleen and liver, but this has never been proven for adults. In this study, we found no patients in the entire cohort had hematopoietic system involvement. For the first time, we demonstrate liver involvement indicates a worse prognosis in adult LCH. The lack of hematopoietic system involvement in this cohort also indicated the different features of the disease in children and adults. In terms of regimen toxicity, grade 3-4 hematological toxicity (mostly neutropenia) was noted in 94% of patients at least once during the treatment. However, the adherence to treatment according to the protocol was high. No patients went off the protocol because of severe adverse events (SAE). Safety benchmarks such as protocol adherence and SAEs confirmed the feasibility of applying MA treatment to adult LCH. Limitations of the current study include this was a single center single arm clinical trial. Since the discovery of recurrent somatic activating mutations of BRAF V600E and additional genetic drivers of the ERK pathway, there have been small studies that demonstrated the efficacy of targeted therapies, BRAF or MEK inhibitors, in LCH.7, 8 Prospective trials should be designed to test the efficacy of targeted therapies in frontline treatment of adult LCH. Also, stratification based on liver involvement at diagnosis may better reflect the clinical diversity. In conclusion, here we report the results of a phase 2 prospective study of newly diagnosed adult MS LCH or SS-m LCH. This study confirmed that methotrexate plus cytarabine was an efficient and feasible regimen with ORR of 87.9%, and 3-year OS and EFS of 97.7% and 68.0%, respectively. The involvement of liver at baseline indicated a worse prognosis in adult LCH. The authors thank the patients and their families. This work was supported by Institutional research funding provided by the Non-profit Central Research Institute Fund of the Chinese Academy of Medical Sciences (2019-RC-HL-001, for CXX), the CAMS Innovation Fund for Medical Sciences (Grant No. 2016-1 2 M-1-002, for LJ) and The National Key Research and Development Program of China (Grant No. 2016YFC0901503, for LJ). The authors declare that they have no conflicts of interest. Cao XX, Li J, Zhou DB and Duan MH contributed to the conception and design of the study; Cao XX, Cai HC, Zhang L, Zhang Y, Feng J and Zhu TN contributed to patient enrolment; Cao XX, Li J, Zhao AL, He TH and Duan MH contributed to data collection; Gao XM and Zhao AL contributed to follow up patients; Cao XX wrote the paper; Niu N assessed all PET/CT evaluations; Sun J and Liang ZY reviewed histological findings independently and all authors revised the paper and approved the submitted version. Table S1. Patient demographic and clinical characteristics Table S2. Risk factors for event-free survival according to univariate and multivariate Cox regression models Table S3. Adverse events 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.

Topics & Concepts