A Risk Model Composed of Complete Blood Count, BRAF V600E and MAP2K1 Predicts Inferior Prognosis of Langerhans Cell Histiocytosis in Children

Background In children, Langerhans cell histiocytosis (LCH), which is the most prevalent histiocytic disorder, exhibits a wide variety of manifestations and outcomes. There is no standard prognosis evaluation system for LCH. We investigated the combined predictive significance of complete blood counts (CBCs), BRAF V600E and MAP2K1 in childhood LCH. Methods A cohort of 71 childhood LCH patients was retrospectively studied. The prognosis predictive significance of platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), systemic inflammation response index (SIRI), systemic immune inflammation index (SII), BRAF V600E, and MAP2K1 were analyzed. Results Histiocyte Society (HS) classification of LCH patients was correlated with NLR, SIRI, and progression free survival (PFS), bone involvement was correlated with SIRI, liver involvement was correlated with NLR, SII, SIRI, and PFS, spleen involvement was correlated with SIRI, lung involvement was correlated with NLR and PFS, CNS involvement was correlated with PFS, while BRAF V600E was correlated with PLR, NLR, SIRI, SII, PFS, and OS (p <0.05). MAP2K1 was correlated with NLR, SIRI, PFS, and OS (p <0.05). Elevated NLR, PLR SIRI, and SII predicted inferior PFS and OS (p <0.05). PLR, NLE, SIRI, SII, BRAF V600E, and MAP2K1 were used to establish a risk model for stratifying the LCH patients into 3 different risk groups. Respective median PFS for low-, mediate-, and high-risk groups were not reached, 26, and 14 months (p <0.001), and all median OS were not reached (p <0.001). Conclusion The risk model combined with CBCs, BRAF V600E, and MAP2K1 might be a promising prognostic system for LCH in children.


INTRODUCTION
Langerhans cell histiocytosis (LCH), also referred to as histiocytosis, defines a collection of histiocytosis disorders of unknown cause. Clinically, LCH exhibits various manifestations, from single indolent lesions to explosive multisystem diseases (1). LCH can develop at any age, with a highest incidence in children aged 1-3 years old. Although the prognoses of LCH patients have slightly improved in the past decades, the standard treatment regimen fails to cure more than 50% patients with the multisystem disease (2). Prognostic prediction depends on whether risk organs (liver, bone, and spleen) are involved at the time of diagnosis, and response to initial therapy (3). However, there is no widely accepted prognosis evaluation system for LCH.
In recent years, the improvement of genomic technology leads to a deep understanding of LCH. In 2010, Badalian-Very et al. (4) reported that about 57% LCH patients had BRAF V600E mutation for the first time. BRAF V600E, a vital kinase of the RAS-RAF-MEK-ERK signal-transduction pathways, has several functions. Apart from BRAF V600E mutations, various activating mutations, such as MAP2K1 mutation, have been reported in LCH (5). Previous studies reported that BRAF V600E mutation may be closely related to multisystem diseases and poorer prognosis (6), and MAP2K1 mutation was related to risk organ involvement (7). With the optimization of LCH treatment regimen, the understanding of genotype needed to be deeper.
Systemic inflammatory responses and host immunity are closely associated with cancer development and prognostic outcomes (8). Complete blood counts (CBC) combined with each other, such as platelet-to-lymphocyte ratio (PLR), neutrophil-to-lymphocyte ratio (NLR), systemic inflammation response index (SIRI), and systemic immune inflammation index (SII), can be used as prognostic predictors for cancers (9,10). However, the significance of CBCs in LCH should be evaluated further.
We investigated the predictive values of CBCs, BRAF V600E, and MAP2K1 in 71 childhood LCH, and evaluated a new risk model that combined with CBCs, BRAF V600E, and MAP2K1 for prognostic prediction of childhood LCH patients.

Study Participants
We used Cochran formula and PASS software to calculate the sample size. Seventy-one childhood LCH patients at the Xiangya Hospital, Central South University were retrospectively evaluated. These patients were diagnosed from January 2014 to December 2020. The inclusion criteria were: i. Pathologically confirmed LCH and ii. Age ≤14 years. The exclusion criteria were: i. Patients administered with radiotherapy, chemotherapy or any other therapies prior to diagnosis; ii. An autoimmune disease history; iii. A chronic inflammatory disease history, including inflammatory bowel diseases; and iv. Uncontrolled infections that are active or the presence of other illnesses.
The 71 LCH patients had their complete follow-up and clinical data. Patient follow-up had been done from the diagnostic day to June 2021, and there was no patient loss to follow-up. The Helsinki Declaration of 1975, which was revised in 2008 was adhered to during this study.

Identification of BRAF V600E and MAP2K1 Mutation
DNA from the paraffin-fixed LCH lesion tissue was extracted by DNA extraction kit (product of QIAGEN, Cat NO.56404) after histologic review by ≥10% histiocytes. BRAF V600E mutations were evaluated by droplet digital polymerase chain reaction (dd-PCR) via the Raindrop system (Raindance Technologies, Billerica, CA) (11). DNA samples were collected from peripheral blood cells, and whole-exome sequencing analysis was performed to detect MAP2K1 mutation status (12).

Data Collection
Clinical data, namely, gender, age, Histiocyte Society (HS) classification, organ involvement, and treatment efficacy after a 6week induced treatment that conducted LCH-III protocolvinblastine and prednisone (13), were obtained. Routine blood analysis was done within a week prior to therapeutic initiation. Various parameters, such as PLR, NLR, SIRI, and SII were determined as: PLR = platelet counts/lymphocyte counts, NLR = neutrophil counts/lymphocyte counts, SIRI = neutrophil counts × monocyte counts/lymphocyte counts, SII = platelet counts × neutrophil counts/lymphocyte counts. A risk model with the above 4 CBC variables, BRAF V600E, and MAP2K1 was evaluated as: i. A score of 1 (low-risk) denoted no mutation of BRAF V600E or MAP2K1, and elevated expression of 0-2 CBCs, ii. A score of 2 (mediate-risk) denotes no mutation of BRAF V600E or MAP2K1, and high expression of 3-4 CBCs; 1 mutation of BRAF V600E or MAP2K1, and high expression of 0-2 CBCs; both of BRAF V600E and MAP2K1 were mutated, and high expression of 0-1 CBCs (3), 1 mutation of BRAF V600E or MAP2K1, and high expression of 3-4 CBCs; both of BRAF V600E and MAP2K1 were mutated, and high expression of 2-4 CBCs. Overall survival (OS) is the time length between diagnostic date to the date of death due to any cause or to last follow-up date. Progression free survival (PFS) denotes the time length from diagnostic date to dates of death or disease progression.
As for the clinico-pathological features of the LCH patients, HS classification was correlated with the levels of NLR and SIRI, bone involvement was correlated with SIRI, liver involvement was correlated with NLR, SII and SIRI, spleen involvement was correlated with SIRI, and lung involvement was correlated with NLR (p <0.05). Gender or age of the patients was not correlated with CBCs (p >0.05; Table 2).
Correlation Between Gene Statuses of BRAF V600E, MAP2K1, and CBCs As shown in Table 2, BRAF V600E mutations were correlated with levels of NLR, SIRI, PLR, and SII, while MAP2K1 mutation was associated with NLR and SIRI (p <0.05).

Prognostic Factors
At follow-up, 31 patients were found to exhibit disease progression. Median PFS was 13 months, 5 patients died, while the median OS was 25 months. As shown in Table 3, HS classification, involvement of liver, lung and CNS were unfavorable PFS predictors (p <0.05), whereas gender, age, HS classification or organ involvement had no correlation with OS (p >0.05).

Prognostic Significance of the Newly Risk Model
Given the predictive values of CBCs, BRAF V600E, and MAP2K1 for LCH, we established a new risk model incorporated NLR, PLR, SII, SIRI, BRAF V600E, and MAP2K1, and evaluated its prognostic value. Table 4 and Figure 4 show that LCH patients were allocated into 3 different groups based on the risk model: low-, mediate-, and high-risk groups. Respective median PFS for low-, mediate-, and high-risk groups were not reached, 26, and 14 months (p <0.001), and all median OS were not reached (p = 0.002).
We analyzed the predictive ability of the risk model in distinguishing treatment efficacy after 6-week treatment.

DISCUSSION
LCH is a histiocytosis disease described by aberrant cellular functions, proliferation or differentiation of mononuclear phagocyte system cells (14). LCH is clinically heterogeneous, ranging from single system involvement to multiple system involvements. The outcome of LCH is highly variable. Therefore, finding prognostic factors for LCH patients is an urgent need. We evaluated the prognostic importance of CBCs, BRAF V600E, and MAP2K1 in childhood LCH, and combined them to build a new risk model. Elevated NLR, PLR, SII, BRAF V600E, SIRI, and MAP2K1 mutations predicted inferior PFS, OS and worse treatment efficacy after a 6-week treatment, and the new risk model could be used for stratifying patients with LCH into various risk categories for prognostic prediction. LCH was initially considered an immune-dysregulatory disorder rather than a neoplastic disorder, until somatic activation of gene mutations in the mitogen-activated protein kinase (MAPK) pathway were found in LCH patients, and reclassified LCH to be a myeloid neoplastic disorder (4,15). About 60% of the LCH patients presented with BRAF V600E mutations, and other genetic alterations that activated the MAPK pathway including MAP2K1 (16). Thus, clinical responses are targeted to BRAF or MEK1 (the product of the MAP2K1 gene) inhibitors which have been observed in LCH with MAP2K1 and BRAF V600E mutations (17). The BRAF V600E mutation is correlated with worse prognostic outcomes in several cancer types. In metastatic colorectal cancers, patients with BRAF V600E mutation had a shorter OS than BRAF V600E wild type (18). In pediatric low-grade glioma, patients harboring the BRAF V600E mutation exhibited poor clinical outcomes after chemotherapy and radiotherapy with a 10-year PFS rate of 27% and 60.2% for the BRAF V600E wild type (19). MAP2K1 mutation is a rare oncogenic alteration, and is reportedly associated with the  (20). In advanced colorectal cancer, all the patients with MAP2K1 mutation treated with anti-EGFR, anti-EGFR combined with MEK and BRAF inhibitors, or anti-EGFR combined with ERK inhibitors, showed disease progression, where the MAP2K1 mutation was associated with poor response to targeted therapy (21). MAP2K1 mutation was also found in splenic diffuse red pulp small B-cell lymphoma, and it was related to an aggressive disease, and the PFS of MAP2K1 mutation patients were shorter than the wild type one (22). In the present study, we found that 26.8% childhood LCH patients presented with BRAF V600E mutations, 7% presented with MAP2K1 mutation, where BRAF V600E and MAP2K1 mutations were predictors of inferior PFS, OS and worse treatment efficacy after a 6-week treatment. Our result was consistent with the previous findings.  factors can affect the tumor microenvironment (TME) (23). Immune cells, such as innate immune cells (neutrophils, NK cells, lymphocytes, dendritic cells, and macrophages) and adaptive immune cells (T and B cells), play important roles in TME. CBCs made up of neutrophils, monocytes lymphocytes, and plates, are reportedly promising in predicting prognosis of several types of cancer. Combined CBCs of the above variates, namely, NLR, SIRI, PLR, and SII are potential markers in prognosis prediction of cancers. In breast cancer, low PLR and NLR are considerably correlated with delayed metastasis, while metastasis-free survival was superior in these patients (24). High preoperative and postoperative NLR, particularly persistently elevated pre-to postoperative NLR are prognostic markers for poor OS in gastric cancer patients (25). In hepatocellular carcinoma, high SII predicted shorter recurrence-free survival and OS (26). In advanced lung adenocarcinoma treated with first-generation EGFR-TKIs, elevated SIRI was correlated with worse ECOG PS, EGFR 19-Del mutation, OS and PFS, SIRI was an independent survival predictor (27). The  Little is known about the effect of these inflammatory cells on LCH.
Our result showed that high NLR, SIRI PLR, and SIRI predicted inferior PFS, OS and worse treatment efficacy after a 6-week treatment, which is consistent with the previous findings.
We established a new risk model that incorporated NLR, PLR, SII, SIRI, BRAF V600E and MAP2K1, and assessed its prognostic significance in childhood LCH. The respective median PFS for low-, mediate-, and high-risk groups were not reached, 26 and 14 months, and all the median OS were not reached. We also found that the risk model could predict treatment efficacy after a 6-week treatment. Feng et al. (28) conducted a CBCs model which consisted of NLR, lymphocyte-monocyte ratio, and PLR. They established that the respective median PFS for low-, intermediate-, and high-risk groups were not reached, 16 and 7 months, while respective median OS for the 3 groups were not reached, 46 and 20 months. Our previous study also established a CBCs model incorporating NLR, PLR, SII, and SIRI for primary central nervous system lymphoma, which accurately predicted the median PFS and median OS (10). Findings of this study are consistent with the previous findings, indicating that the new risk model could be used for prognosis prediction in childhood LCH.
In summary, our study showed that the new risk model combined with CBCs, BRAF V600E, and MAP2K1 might be a promising prognostic system for LCH in children. However, some limitations associated with the present study are worth mentioning. First, we enrolled a small sample size at a single center. Second, the CBCs can be influenced by a variety of factors, including immune status that may bias the findings. Studies should be conducted to verify the roles of CBCs, BRAF V600E, and MAP2K1 in childhood LCH.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the Ethics Committee of Xiangya Hospital of Central South University. Written informed consent to participate in this study was provided by the participants' legal guardian/next of kin.

AUTHOR CONTRIBUTIONS
LW and CL designed the research study and collected the clinical data. LinY, XD, KZ, YY performed the research. QQ, LiaY and YX and drafted the manuscript. XQu, HL, and XQi participated in the literature search and analyzed the data. All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

FUNDING
The study was supported by funds from the NSFC under the following grant numbers: #81600026, #31900424, #81970033, #82070034; funding from the Open Foundation of Hunan College Innovation Program (grant number #20K142), funds from the Hunan Natural Science Foundation under grant numbers #2019JJ50760, #2020JJ4688, and #2020JJ4776, and funding from the Open Sharing Fund for the Lager-scale Instruments and Equipment of Central South University.

ACKNOWLEDGMENTS
Much appreciation is given to all the staff of the Xiangya Hospital, Central South University and the Department of Physiology, Xiangya School of Basic Medicine Science, Central South University, who supported this study.