eTable 1. Multivariate Cox Regression Analysis of the Association Between Minimum ALC and Overall Survival
eTable 2. Multivariate Cox Regression Analysis of the Association Between Minimum ALC, Stratified by Lymphopenia Grade, and Overall Survival
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Brewster R, Purington N, Henry S, Wood D, Ganjoo K, Bui N. Evaluation of Absolute Lymphocyte Count at Diagnosis and Mortality Among Patients With Localized Bone or Soft Tissue Sarcoma. JAMA Netw Open. 2021;4(3):e210845. doi:10.1001/jamanetworkopen.2021.0845
Is there an association between lymphopenia at diagnosis and overall mortality among patients with localized bone and soft tissue sarcoma?
In this cohort study of 634 patients with localized bone and soft tissue sarcoma, peripheral lymphocyte counts at diagnosis were significantly inversely associated with overall mortality.
This study suggests that absolute lymphocyte count may constitute a reliable prognostic biomarker for patients with bone and soft tissue sarcoma; however, further research is needed to define the mechanistic role of host antitumor immunity and infectious complications.
Host-related immune factors have been implicated in the development and progression of diverse malignant neoplasms. Identifying associations between immunologic laboratory parameters and overall survival may inform novel prognostic biomarkers and mechanisms of antitumor immunity in localized bone and soft tissue sarcoma.
To assess whether lymphopenia at diagnosis is associated with overall survival among patients with localized bone and soft tissue sarcoma.
Design, Setting, and Participants
This retrospective cohort study analyzed patients from the Stanford Cancer Institute with localized bone and soft tissue sarcoma between September 1, 1998, and November 1, 2018. Patients were included if laboratory values were available within 60 days of diagnosis and, if applicable, prior to the initiation of chemotherapy and/or radiotherapy. Statistical analysis was performed from January 1, 2019, to November 1, 2020.
Absolute lymphocyte count within 60 days of diagnosis and antimicrobial exposure, defined by the number of antimicrobial agent prescriptions and the cumulative duration of antimicrobial administration within 60 days of diagnosis.
Main Outcomes and Measures
The association between minimum absolute lymphocyte count at diagnosis and 5-year overall survival probability was characterized with the Kaplan-Meier method and multivariate Cox proportional hazards regression models. Multivariable logistic regressions were fitted to evaluate whether patients with lymphopenia were at greater risk of increased antimicrobial exposure.
Among 634 patients, the median age at diagnosis was 53.7 years (interquartile range, 37.5-66.8 years), and 290 patients (45.7%) were women, with a 5-year survival probability of 67.9%. There was a significant inverse association between lymphopenia at diagnosis and overall survival (hazard ratio [HR], 1.82; 95% CI, 1.39-1.40), resulting in a 13.5% 5-year survival probability difference compared with patients who did not have lymphopenia at diagnosis (60.2% vs 73.7% for those who never had lymphopenia). In addition, poorer survival was observed with higher-grade lymphopenia (grades 3 and 4: HR, 2.44; 95% CI, 1.68-3.55; grades 1 and 2: HR, 1.60; 95% CI, 1.18-2.18). In an exploratory analysis, patients with increased antibiotic exposure were more likely to have lymphopenia (odds ratio, 1.96; 95% CI, 1.26-3.07 for total number of antimicrobial agents; odds ratio, 1.70; 95% CI, 1.10-2.57 for antimicrobial duration) than antimicrobial-naive patients.
Conclusions and Relevance
This study suggests that an abnormally low absolute lymphocyte count at diagnosis is associated with higher mortality among patients with localized bone and soft tissue sarcoma; therefore, lymphopenia may serve as a reliable prognostic biomarker. Potential mechanisms associated with host immunity and overall survival include a suppressed antitumor response and increased infectious complications, which merit future investigation.
Sarcomas constitute a rare but heterogenous group of malignant neoplasms originating from mesenchymal tissue. Despite advances in diagnostic and therapeutic modalities, outcomes among patients with high-grade sarcoma have remained poor.1 It is critical to validate novel biomarkers to stratify high-risk patients who may benefit from earlier aggressive therapeutic regimens.
The immunosurveillance theory posits that lymphocytes play a protective role in detecting and targeting malignant cells.2,3 This theory has generated substantial interest in developing immunotherapies for sarcomas, with promising results among a subset of patients.4 Furthermore, reports indicate that lymphopenia before treatment initiation is associated with higher mortality across diverse cancer types.5
In this retrospective cohort study of 634 patients treated at the Stanford Cancer Institute in Stanford, California, between 1998 and 2017, we evaluated the prognostic value of lymphopenia at diagnosis for patients with localized bone and soft tissue sarcoma and assessed the association of infectious complications as measured by antimicrobial exposure with lymphopenia.
We identified 634 eligible patients with a diagnosis of bone or soft tissue sarcoma who received a diagnosis at the Stanford Cancer Institute between September 1, 1998, and November 1, 2018, using a retrospective analysis within the Stanford Cancer Institute Research Database. Patients were included if they received a formal diagnosis of sarcoma at Stanford Cancer Institute and had laboratory values measured within 60 days of date of diagnosis. Patients with metastatic disease at diagnosis were excluded. This study was approved by the Stanford University institutional review board, which granted a waiver of consent because the research involved no more than minimal risk to participants and procedures were in place to protect confidentiality. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
Patients were classified as ever having lymphopenia if they had an absolute lymphocyte count (ALC) value less than 1000/µL (to convert to ×109 cell per liter, multiply by 0.001) within 60 days of the date of sarcoma diagnosis and before the start of chemotherapy and/or radiotherapy. The minimum ALC refers to the lowest ALC value among all those measured during the aforementioned timeline. For patients designated as ever having lymphopenia, minimum ALC was graded in accordance with the Common Terminology Criteria for Adverse Events.6
We evaluated the association between use of antimicrobial agents and lymphopenia. Antibiotics and antifungals of interest were considered if orders were recorded within 60 days of diagnosis and prior to the initiation of chemotherapy and/or radiotherapy, if given. Exposure was quantified as both (1) the total number of unique antibiotics and antifungal prescriptions and (2) the total duration in nonoverlapping days receiving antibiotics and antifungals. Any patient who did not survive beyond 150 days after diagnosis was excluded from analysis. For patients receiving antimicrobials, median values were used as thresholds to delineate the subgroups (0, 1-3, and ≥4 antimicrobial agents and 0, 1-3, and ≥4 total days of antimicrobial administration).
Statistical analysis was performed from January 1, 2019, to November 1, 2020. The Kaplan-Meier method was used to estimate survival curves stratified on the basis of lymphopenic status or antibiotic exposure. Multivariate Cox proportional hazards regression models were generated to evaluate the association between minimum ALC and overall mortality. We adjusted for potential explanatory variables, including minimum white blood cell count, minimum absolute neutrophil count, tumor grade, and age at diagnosis. The asociations between lymphopenia and antibiotic exposure was assessed with multivariate logistic regression models, adjusted for tumor grade and age at diagnosis. All statistical analyses were performed with the use of R, version 1.1.463. P = .05 was defined as the threshold for statistical significance.
Between 1998 and 2017, a total of 2240 patients received a diagnosis of sarcoma at the Stanford Cancer Center, and 634 met the inclusion criteria. The median age at diagnosis was 53.7 years (interquartile range [IQR], 37.5-66.8 years), and 290 patients (45.7%) were women (Table 1). At diagnosis, the median value for ALC was 1120/µL (IQR, 700-1610/µL), the median white blood cell count was 6600/µL (IQR, 5200-7980/µL [to convert to ×109 cells per liter, multiply by 0.001]), and the median absolute neutrophil count was 4610/µL (IQR, 3490-6190/µL [to convert to ×109 cells per liter, multiply by 0.001]). A total of 281 patients (44.3%) ever had lymphopenia within 60 days of diagnosis. Five-year survival probability was 67.9%. The most common histologic subtypes were pleomorphic sarcoma (121 [19.1%]), liposarcoma (94 [14.8%]), and chondrosarcoma (65 [10.3%]).
The minimum ALC ranged from 60 to 5490/µL (median, 1120/µL). On multivariate analysis, patients who had ever had lymphopenia had a worse survival after adjusting for minimum white blood cell count, minimum absolute neutrophil count, tumor grade, and age at diagnosis (hazard ratio [HR], 1.82; 95% CI, 1.39-2.40) (Figure, A, and eTable 1 in the Supplement). This corresponded to a 60.2% 5-year survival probability compared with 73.7% for patients who never had lymphopenia. Grade 3 or 4 lymphopenia were associated with poorer outcomes (HR, 2.44; 95% CI, 1.68-3.55) than grade 1 or 2 lymphopenia (HR, 1.60; 95% CI, 1.18-2.18) (Figure, B, and eTable 2 in the Supplement). There was no significant association between ever having leukopenia (HR, 0.61; 95% CI, 0.33-1.12) or ever having neutropenia (HR, 0.58; 95% CI, 0.20-1.68) and survival.
A total of 302 of 602 patients (50.2%) did not receive any antimicrobials (Table 2). A total of 164 of 602 patients (27.2%) received 1 to 3 antimicrobial agents, and 136 of 602 patients (22.6%) received 4 or more distinct antimicrobial agents. A total of 163 of 602 patients (27.1%) had 1 to 3 days of antimicrobial exposure, and 133 of 602 patients (22.1%) had 4 or more days of antimicrobial exposure. There was no association between any amount of antimicrobial exposure and lymphopenia (odds ratio, 1.15; 95% CI, 0.82-1.59). However, on subgroup analysis, patients with 4 or more antimicrobial regimens were significantly more likely than antimicrobial-naive patients to have lymphopenia (odds ratio, 1.96; 95% CI, 1.26-3.07) (Table 2). Furthermore, patients whose cumulative duration of antimicrobial treatment was 4 or more days were also significantly more likely to have lymphopenia (OR, 1.70; 95% CI, 1.10-2.57)
To our knowledge, this is the first report to identify lymphopenia at diagnosis of localized sarcoma as an independent prognostic biomarker. Among sarcomas, lymphopenia prior to initiating systemic therapy has been shown to correlate with poor survival among patients with advanced disease.5 Similarly, increased rates of recurrence and mortality after surgery are observed among patients with preoperative lymphopenia.7 Antitumor immunity relies on the recruitment of specific immune compartments to the tumor microenvironment. It is subject to complex autoregulation by proinflammatory molecules (eg, tumor-infiltrating lymphocytes) and anti-inflammatory molecules (eg, regulatory T cells and tumor-associated macrophages). The detection of tumor-infiltrating lymphocytes in retrospective histopathologic analysis corresponds to improved prognosis in various solid tumors and sarcoma subtypes.8-10 Pending further mechanistic insights, it is possible that a low peripheral lymphocyte count at sarcoma diagnosis limits the ability to mount a durable antitumor response, which may be associated with the poorer outcomes observed among patients with lymphopenia.
Our exploratory analysis of antimicrobial exposure suggests another possible consequence of lymphopenia. Opportunistic infections are a principal cause of mortality among patients with cancer.11 They have been associated with lymphocyte depletion, typically in the context of treatment-associated toxic effects.12,13 For example, lymphopenia secondary to chemotherapy in breast cancer has been shown to significantly increase the risk for Pneumocystis pneumonia.14-16 Thus, the poor outcomes observed among patients with sarcoma with lymphopenia at diagnosis may be partially explained by higher rates of infectious complications even prior to the initiation of therapy.
There are limitations to our study. First, this was a single-center observational cohort study; larger-scale studies are needed to assess the generalizability of our findings. In addition, we were unable to control for all possible confounding variables, and complete laboratory and clinical features were not available across the study population. There were also limitations with the method for evaluating antimicrobial exposure. Without hospital admission data, we approximated the burden of infections on the basis of antibiotic prescribing patterns, which do not exclude prophylactic indications or accurately capture the severity of infections. Ultimately, any suggestion that antibiotic exposure is associated with lower survival would need to be explored in future analyses.
The findings of this study suggest that the ALC at diagnosis is inversely associated with mortality among patients with localized sarcoma and may potentially be used as a prognostic biomarker. Future studies should further evaluate clinical mediators of lymphopenia as well as the correlation between peripheral ALC and recruitment of tumor-infiltrating lymphocytes and other immune compartments to the tumor microenvironment. Altogether, these results add to the emerging understanding of host immunity in sarcoma and other solid tumor malignant neoplasms and suggest the powerful role of the native immune system in regulating cancer progression.
Accepted for Publication: January 17, 2021.
Published: March 5, 2021. doi:10.1001/jamanetworkopen.2021.0845
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Brewster R et al. JAMA Network Open.
Corresponding Author: Nam Bui, MD, Department of Medicine (Oncology), Stanford University School of Medicine, 780 Welch Rd, Room CJ250J, Palo Alto, CA 94304 (email@example.com).
Author Contributions: Dr Bui had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Brewster, Ganjoo, Bui.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Brewster, Bui.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Brewster, Purington, Ganjoo, Bui.
Obtained funding: Bui.
Administrative, technical, or material support: Wood, Bui.
Supervision: Purington, Ganjoo, Bui.
Conflict of Interest Disclosures: None reported.