ESTRO 2021 Abstract Book

S1241

ESTRO 2021

Conclusion Weekly assessment of stress was successfully introduced during the COVID-19 pandemic, identified profession and workplace specific stress patterns and allowed precise counteractions. Whether these measures reduced stress levels needs to be further evaluated. PO-1515 Meta-analysis on the influence of radiation-induced lymphopenia on overall survival in solid tumors. P. Damen 1 , T. Kroese 1 , R. van Hillegersberg 2 , E. Schuit 3 , M. Peters 1 , J. Verhoeff 1 , S. Lin 4 , P. van Rossum 1 1 University Medical Center Utrecht, Department of Radiation Oncology, Utrecht, The Netherlands; 2 University Medical Center Utrecht, Department of Surgery, Utrecht, The Netherlands; 3 University Medical Center Utrecht, Utrecht, Julius Center for Health Sciences and Primary Care, Utrecht, The Netherlands; 4 The University of Texas MD Anderson Cancer Center, Department of Radiation Oncology, Houston, USA Purpose or Objective Emerging evidence suggests an independent detrimental prognostic association between radiation-induced lymphopenia (RIL) and pathologic response, progression-free and overall survival in patients who undergo radiotherapy for cancer. The aim of this study was to systematically review and meta-analyze the prognostic impact of RIL on overall survival (OS) in patients with solid tumors. Materials and Methods PubMed/MEDLINE and Embase were systematically searched. The analysis included intervention and prognostic studies that reported on the prognostic relationship between RIL and survival in patients with solid tumors. An overall pooled adjusted hazard ratio (aHR) was calculated using a random-effects model. Subgroup analyses for different patient-, tumor-, treatment- and study-related characteristics were performed using meta- regression. Lymphopenia was graded according to the Common Terminology Criteria for Adverse Events version 5.0. Results Twenty-one cohorts within 20 studies reporting on adjusted HRs were eligible for meta-analysis. Radiotherapy target sites included tumors of the brain, lung, pancreas, esophagus, liver, cervix, oropharynx, bone, and anal canal. Pooled analysis across the 16 cohorts reporting on grade ≥3 RIL (versus grade 0-2 RIL) demonstrated a statistically significant detrimental OS with an overall adjusted pooled HR of 1.65 (95% confidence interval [CI]: 1.43-1.90), as demonstrated in Figure 1. Moderate heterogeneity among HRs was observed (I 2 =49.1%), mostly attributable to overestimated HRs in 7 studies which were likely subject to model-overfitting. Subgroup analysis per tumor site showed a statistically significant association between grade ≥3 RIL and worse OS in 4 brain tumor studies (pooled aHR 1.63, 95% CI: 1.06-2.51; I 2 =3.6%), 4 lung cancer cohorts (pooled aHR 1.52, 95% CI: 1.01-2.29; I 2 =0%), and 3 pancreatic cancer studies (pooled aHR 1.92, 95% CI: 1.10-3.36; I 2 =0%). In addition, pooled analysis across 5 studies reporting on grade 4 RIL (versus grade 1-3 RIL) showed a statistically significant pooled aHR for detrimental OS of 1.53 (95% CI: 1.24-1.90; I 2 =29.1%), as demonstrated in Figure 2. Figure 1. Forest plot of the pooled analysis of the prognostic value of grade ≥3 RIL on OS.