Association between body mass index and outcomes in lymphoma‐associated haemophagocytic lymphohistiocytosis: A retrospective multicentre cohort study of Jiangsu Cooperative Lymphoma Group ( JCLG )

Haemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyperinflammatory syndrome caused by dysregulation of cytotoxic T cells and macrophage activation. Among the subtypes of HLH, lymphoma-associated HLH (LA-HLH) carries a worse prognosis compared to HLH caused by infections or autoimmune diseases.1 Previous studies have shown that the median survival for patients with LA-HLH is only about 2 months.2 It is essential to identify a robust marker to enable better risk stratification and guide therapeutic decisions in LA-HLH. Body mass index (BMI) is a standard measurement used to classify weight status. However, its role in predicting cancer outcomes remains controversial.3-5 Critically, the impact of BMI on outcomes specifically in LA-HLH is unknown. Therefore, we conducted this multicentre retrospective study to explore the prognostic impact of BMI in patients with LA-HLH. This multicentre retrospective study included LA-HLH patients from 25 medical centres in the Jiangsu Cooperative Lymphoma Group (JCLG) diagnosed between 2017 and 2024. Data on weight and height on the first day of the index hospitalization were acquired. BMI was calculated as weight (kg) divided by height squared (m2). Patients were categorized according to Chinese BMI criteria6, 7: underweight (BMI < 18.5 kg/m2), normal weight (BMI: 18.5–23.9 kg/m2), overweight (BMI: 24.0–27.9 kg/m2) and obese (BMI ≥28.0 kg/m2). Analyses were performed using R software (v4.4.2) with the survival, purrr, riskRegression and survminer packages. A two-sided p < 0.05 was considered statistically significant. The clinical characteristics of the 315 LA-HLH patients are summarized in Table 1. The median BMI of the cohort was 22.48 kg/m2 (IQR: 20.39–24.32). Based on Chinese BMI criteria, 34 patients (10.8%) were underweight, 192 (60.9%) were normal weight, 74 (23.5%) were overweight and 15 (4.8%) were obese. If not specified, BMI groups in this study referred to subgroups stratified by Chinese BMI criteria. Among the cohort, 223 patients had died and 92 remained alive, with a median follow-up duration of 1162 days. The 60-day and 365-day survival rates were 61.89% and 39.53% respectively. Underweight, overweight or obese patients had worse 60-day survival and OS than normal weight patients (Figure 1A,B). Survival analysis based on the WHO BMI classification yielded results consistent with those from the original BMI groups. Specifically, Kaplan–Meier curves showed a significant difference in 60-day survival (p = 0.048, Figure S1A), whereas the difference in OS was not statistically significant (p = 0.064, Figure S1B). Due to the limited number of obese patients, we combined obese and overweight patients into an overweight/obese group. Univariate analysis identified several factors significantly associated with inferior 60-day survival and OS (Table S1). Multivariate analysis demonstrated that overweight/obese status independently predicted worse 60-day survival (p = 0.010, Figure 1C), however, its independent role in predicting OS was only marginally significant (p = 0.062, Figure 1D). Additionally, T/natural killer (NK)-cell lymphoma, composite lymphoma, refractory/relapse (RR) disease and Eastern Cooperative Oncology Group performance status (ECOG PS) >2 were identified as independent adverse prognostic factors for both 60-day survival and OS (Figure 1C,D). Regarding the initial treatments, 219 (69.5%) received HLH-directed therapy, 85 (27.0%) lymphoma-specific therapy and 11 (3.5%) received neither. A total of 250 patients received lymphoma-specific therapy with or without prior HLH-directed therapy, with 111 (44.4%) receiving etoposide-containing lymphoma-specific therapy. Initial treatment patterns did not differ significantly across BMI groups (p = 0.840; Figure S2A) nor did the incorporation of etoposide into lymphoma-specific therapy (p = 0.072; Figure S2B). In patients receiving HLH-directed therapy as initial treatment, both overweight/obese and underweight groups tended to have worse 60-day survival compared to normal weight patients, although this difference was not statistically significant (p = 0.065; Figure 1E). No significant difference was observed in OS (p = 0.210; Figure S2C). Among patients receiving lymphoma-specific therapy as initial treatment, BMI showed significant prognostic impacts on 60-day survival and OS (Figure 1F and Figure S2D). BMI showed no prognostic significance in patients treated with etoposide-containing lymphoma-specific therapy (Figure S2E,F). In contrast, among those receiving non-etoposide regimens, both overweight/obese and underweight patients had significantly worse survival outcomes than normal weight patients (Figure S2G,H). No significant survival difference was observed between etoposide-containing and non-etoposide-containing lymphoma-specific therapy in normal weight patients (Figure 1G and Figure S3A). In contrast, overweight/obese patients treated with etoposide-containing lymphoma-specific therapy exhibited significantly better 60-day survival and OS than those receiving non-etoposide regimens (Figure 1H and Figure S3B). Similarly, our analysis revealed a trend towards improved survival in underweight patients treated with etoposide-containing lymphoma therapy, although without statistical significance (Figure S3C,D). Additionally, we analysed the prognostic impacts of incorporation of etoposide into lymphoma-directed therapy in subgroups stratified by several other parameters. In T/NK-cell lymphoma-associated HLH, the incorporation of etoposide into lymphoma-directed therapy was significantly associated with improved 60-day survival (p = 0.022, Figure S4A) and OS (p = 0.002, Figure S4B). In B-cell lymphoma-associated HLH, the incorporation of etoposide into lymphoma-directed therapy improved 60-day survival (p = 0.013, Figure S4C), but not OS (p = 0.650, Figure S4D). In patients aged ≤ 60 years, lymphoma-specific therapy incorporating etoposide was associated with significantly improved 60-day survival (p < 0.001; Figure S4E), whereas no significant benefit was observed for OS (p = 0.130; Figure S4F). Conversely, in patients aged > 60 years, the incorporation of etoposide into lymphoma-specific therapy did not confer a significant survival advantage in either 60-day survival (p = 0.570; Figure S4G) or OS (p = 0.840; Figure S4H). Furthermore, a subgroup analysis based on ECOG PS (ECOG PS≤2 vs. ECOG PS>2) was performed in patients aged ≤60 years. Among patients with ECOG PS≤2, those treated with etoposide-containing lymphoma-specific therapy had significantly better 60-day survival (p = 0.008, Figure S5A), though no difference in OS was observed (p = 0.240, Figure S5B). Similarly, patients with ECOG PS>2 receiving etoposide-containing lymphoma-specific therapy had better 60-day survival (Figure S5C). However, incorporation of etoposide showed no impacts on OS of patients with ECOG PS>2 (Figure S5D). Age-stratified analysis revealed significant heterogeneity in BMI's prognostic impact. In patients ≤60 years, underweight and overweight/obese patients had worse 60-day survival and OS compared to normal weight patients (Figure S6A,B). In contrast, BMI had no effect on survival in patients >60 years (Figure S6C,D). Multivariable models showed that, among younger patients, overweight/obese status independently predicted worse 60-day survival (hazard ratio [HR] = 2.35, 95% CI 1.20–4.63; p = 0.013) and OS (HR = 1.64, 95% CI 1.03–2.62; p = 0.039) (Figure 1I,J). However, BMI did not significantly impact prognosis in older patients (all p > 0.05). Obesity is a recognized risk factor for death and relapse in cancer patients, while the impact of BMI on the prognosis of patients with LA-HLH has not been reported. This study retrospectively analysed the influence of BMI at HLH diagnosis on prognosis in 315 patients with LA-HLH. We found that overweight or obesity at diagnosis was associated with poor prognosis specifically in patients aged ≤60 years with LA-HLH. In this study, overweight/obese patients had significantly poorer outcomes than normal weight patients. Obesity promotes tumour progression through mechanisms such as the production of adipokines that drive tumour growth, metastasis and remodelling of the tumour microenvironment.8 Furthermore, previous studies have shown that adipocytes can promote tumour progression by stimulating the proliferation of tumour stem cells.9 Adipose tissue may increase the severity of systemic inflammatory response by secreting pro-inflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6), thereby leading to poor prognosis for the patients. Park et al.10 have shown that obesity leads to changes in the body's endocrine system and creates a unique microenvironment of adipose tissue. Together, these alterations promote both the initiation and progression of tumours. Therefore, we speculate that, in LA-HLH, obesity may both exacerbate the inflammatory storm and promote lymphoma progression, thereby affecting patient prognosis. Etoposide is widely recognized as a cornerstone therapy for HLH. The study by Zoref-Lorenz et al.11 demonstrates improved survival in LA-HLH patients receiving etoposide-containing lymphoma regimens. Our study shows that overweight/obese patients receiving etoposide-containing lymphoma-specific therapy had better survival, while normal weight patients did not benefit from incorporation of etoposide into lymphoma-specific therapy. These findings need to be confirmed in future studies. In this study, we observed that the adverse prognostic impact of elevated BMI is only in younger patients (≤60 years), not in older patients (>60 years). This might be because the elderly population has a higher comorbidity burden and diminished physiological reserve,12 which weakens the predictive value of BMI. As a retrospective analysis, even after statistical adjustments, our research still has inherent limitations, such as potential selection bias and unmeasured confounding factors. Furthermore, it is particularly challenging to clearly determine the causes of death, which complicates the interpretation of the results. Finally, complications were not systematically recorded in our records, especially for elderly patients. In future studies, we will further improve this aspect. This study identifies overweight/obese at diagnosis as an independent adverse prognostic factor for younger patients (≤60 years) with LA-HLH. It is worth noting that the age-dependent nature of BMI highlights the importance of personalized risk assessment. In the future, it is necessary to further validate the prognostic role of BMI through prospective studies. The survival benefit of etoposide-containing regimens in overweight/obese patients warrants further investigation in future studies. Jianyong Li, Bingzong Li, Yi Miao and Wenyu Shi conceptualized and supervised this study. Yanping Liu and Yi Miao analysed the data and wrote the original draft. Jing Zhang, Ling Wang, Zihan Liang, Yingyun Gong, Ling Gao, Wei Zhang, Xuzhang Lu, Chunling Wang, Liang Yu, Miao Sun, Yun Zhuang, Yuqing Miao, Haiwen Ni, Xiaoyan Xie, Jingyan Xu, Yunping Zhang, Min Zhao, Min Xu, Wanchuan Zhuang, Weiying Gu, Guoqiang Lin, Haiying Hua, Jianfeng Zhu, Maozhong Xu, Tao Jia, Lijia Zhai, Huirong Shan, Qiudan Shen and Lei Fan collected the data. All authors reviewed the manuscript and approved the final draft of the manuscript. We acknowledge all the contributors not listed as authors in this study. The authors declare no competing interests. This work was supported by Clinical Capacity Enhancement Project (JSPH MC-2023-6). This study protocol received approval from the Ethics Committee of the Affiliated Hospital of Nantong University. Patient consent was waived because this was a retrospective and deidentified study. The data that support the findings of this study are available on request from the corresponding author. Data S1. 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.