Vienna, Austria

ESTRO 2023

Session Item

Biomarkers
6015
Poster (Digital)
Clinical
Dynamic neutrophil-to-lymphocyte ratio - a new predictive biomarker for recurrent sarcomas?
Constanza Martinez, Canada
PO-1444

Abstract

Dynamic neutrophil-to-lymphocyte ratio - a new predictive biomarker for recurrent sarcomas?
Authors:

Constanza Martinez1, Fabio Cury1, Carolyn Freeman1, Rie Asso1, Neelabh Rastogi2

1McGill University Health Centre, Radiation Oncology, Montreal, Canada; 2McGill University, Faculty of Medicine and Health Sciences, Montreal, Canada

Show Affiliations
Purpose or Objective

The neutrophil-to-lymphocyte ratio (NLR) is an accessible measure of systemic inflammation known to have a prognostic value across different cancers. In sarcomas, NLR has been mainly studied as a prognostic marker of response to chemotherapy and targeted therapy. Our study aims to determine if dynamic changes in NLR from diagnosis and throughout radiation therapy (RT) treatment are associated with the time of recurrence. Here we sought to determine the value of NLR as an independent predictive and prognostic biomarker in adult soft tissue sarcomas (STS).

Material and Methods

We conducted a retrospective analysis for recurrent sarcomas (RS) and a paired number of non-recurrent sarcoma (NRS) patients, treated between 2013 and 2022. Clinical and demographic data were collected, including radiation therapy treatments, and complete blood count (CBC). NLR was calculated by dividing the absolute neutrophils by absolute lymphocytes. The NLR was calculated for RS at 4-time points: diagnosis, post-radiation therapy (RT), recurrence, and after a  second-RT treatment. For NRS the NLR was calculated at diagnosis and post-RT. We used correlation analyses, t-test and simple proportions for data analyses.

Results

We identified 28 patients that presented with one or more recurrences. Of all patients, 9 had available CBCs results at the 4 selected time points. RS numbers were compared with 9 NRS patients. The median age at diagnosis for NRS was 67.3 years and for RS 65.3 years. The histology in NRS and RS were respectively; Myxofibrosarcoma n=2, n=2), spindle cell (n=0, n =3), undifferentiated pleomorphic sarcoma (n=1, n=2), leiomyosarcoma (n=2, n=1), myxoid liposarcoma (n=2, n=1), and other (n=2, n=0). For RS the mean NLR at diagnosis was 4.14 (95%CI= 3.12-5.15), at post-RT 3.72 (95%CI=2.52-4.92), at recurrence 8.43 (95%CI= 3.86-13.01) and at the time of second-RT 5.39 (95%CI= 2.91-7.88). For NRS the mean NLR at diagnosis was 2.55 (95%CI= 0.31-1.48) and post-RT 4.29 (95%CI= (0.66-3.15). The delta changes in NLR (dNLR) between diagnosis and RT for RS was 0.99 (95%CI= 0.11-1.83) and for NRS was 1.73 (95%CI= 1.14-8.66). The baseline diagnostic NLR in RS and NRS was not correlated with histology or age. The baseline diagnostic NLR among both RS and NRS showed a significant correlation between tumor and higher NLR (t-test, p= 0.0108). In the RS group, 2 patients died of disease. These patients had the highest dNLR between post-RT and recurrence above 3.7.

Conclusion

A higher baseline NLR at diagnosis is associated with worse local control in STS and with larger tumor at diagnosis. Dynamic changes in NLR for RS can predict better outcomes and detect the time of recurrence. A high dNLR between diagnosis and RT is associated with better outcomes. Despite the small number of patients, this is the first study to assess the dynamic changes in NLR and its association with prognosis and prediction of recurrence in STS.