ESTRO 2021 Abstract Book

S720

ESTRO 2021

Conclusion First order feature variance might be able to predict PFS for patients with esophageal cancer who were treated with dCRT. FDG-PET/CT radiomics is useful for esophageal cancer.

PD-0883 Impact of unintentional splenic scatter on lymphocytes during and post stomach cancer chemoradiation Y.H.E. Chuk 1 , K.C.M. Cheung 1 , J.C.H. Chow 1 , K.M. Cheung 1 , J.C.H. Chan 1 , C.W.Y. Yip 1 , F.K.H. Lee 1 , K.H. Wong 1 1 Queen Elizabeth Hospital, Clinical Oncology, Hong Kong, Hong Kong (SAR) China Purpose or Objective The spleen has been known as a paramount hematological organ in human body immunity. A small dose of radiation is sufficient to affect lymphoid hematopoiesis. However, scatter radiation dose to the spleen is still largely unregulated in most radiation planning procedures. This study aims to investigate the impact of unintentional splenic scatter on serial absolute lymphocyte counts (ALC) during and after adjuvant chemoradiation in stomach cancer patients. Materials and Methods A cohort of consecutive stomach cancer patients who received adjuvant chemoradiation from Jan 2010 to Mar 2020 was recruited in a single tertiary cancer institution. Patients with baseline active autoimmune diseases, HIV infection, history and synchronous solid and hematological malignancy apart from stomach cancer, incomplete treatment and receiving palliative chemotherapy during the time of post treatment serial blood taking were excluded. Baseline demographics including age, histology, charlson comorbidity index, AJCC staging (8 th edition), tumour grade, chemotherapy intensity, treatment interruption, baseline and serial complete blood counts at baseline, week 5, week 12, week 24, week 52 were collected. Splenic dosimetric parameters including maximum splenic radiation dose (Dmax), mean splenic radiation dose (Dmean), V1, V3, V5, V10, V15, V20, V30, V40, V45 (volume receiving ≥ X Gy in cc) and splenic volume were collected. EQD2 was calculated using alpha/beta of 3 for the spleen. Serial ALC were dichotomized into <0.8x10 9 /L and ≥ 0.8 x10 9 /L based on Grade 2 lymphopenia and above according to the Common Terminology Criteria for Adverse Events version 5.0. Mann-Whitney U test was used to look for association between Vx and dichotomised ALC. P-value <0.05 was considered statistically significant. Results 58 eligible patients were analysed. Median age was 59 years old, majority were smoker (25/58), non-drinker (48/58), charlson comorbidity index of 0-1, adenocarcinoma histology (57/58), grade 3 tumor (33/58), and stage III disease (33/58). Median ALC at week 0, 5, 12, 24, 52 were 1.8x10 9 /L, 0.7x10 9 /L, 0.4x10 9 /L, 0.9x10 9 /L and 1.01 x10 9 /L respectively. Median splenic volume was 112.35cc. Higher splenic dosimetric parameters V1, V3, V5, V10, V15, V20, V30, V40, V45 (cc) were found to be associated with lower ALC at week 5 (all p-values <0.05). However, no significant difference in splenic dosimeter parameters was observed between low and high ALC at week 12, 24 and 52. No patients with spleen V1 < 100cc has ≥ grade 2 lymphopenia at week 5. Conclusion Unintentional splenic scatter appears to affect short term ALC in stomach cancer patients receiving adjuvant chemoradiation while long term ALC were unaffected. Splenic constraints should be considered to reduce acute hematological side effects in stomach cancer patients who undergo chemoradiation.