Vienna, Austria

ESTRO 2023

Session Item

Tumour radiobiology
Poster (Digital)
Radiosensitivity Index as potential marker of biological sex differences to radiation response
Laure Marignol, Ireland


Radiosensitivity Index as potential marker of biological sex differences to radiation response

Laure Marignol1, David Robert Grimes2

1Trinity College Dublin , Radiation Therapy, Dublin , Ireland; 2Trinity College Dublin, Radiation Therapy, Dublin , Ireland

Show Affiliations
Purpose or Objective

Sex is a biological variable increasingly reported to impact cancer treatment outcomes. The purpose of this in silico study was to investigate whether biological sex is associated with differences in genomic scores defining radiosensitivity; and test the hypothesis that the associated genomic-adjusted radiation dose will reveal inter-sex differences.  

Material and Methods

Freely available, raw RNA-Seq and gene chip data were accessed from the Cancer Gene Atlas for the 10 Radiosensitivity Index (RSI) genes (AR, cJun [JUN], STAT1, PKC-β [PRKCB], RelA [RELA], cABL [ABL], SUMO1, PAK2, HDAC1, and IRF1) across 4 main cancer sites: head and neck  (N= 364 male, N=133 female), bladder (N=  291 male, N=106 female), rectum (N= 90 male, N=75 female) and lung (N=1014 male, N=707 female). The RSI was calculated using the published equations for each patient. The RSI value was subsequently used to derive the Genomically adjusted radiation dose (GARD) score for each patient. To test differences between the sexes, two tailed independent student t-tests were used.


In male cohorts, the mean ± SD RSI values were 0.33±0.17 (lung). 0.45± 0.21 (head and neck), 0.5±0.19 (rectal) and 0.14±0.21 (bladder). In female cohorts, the mean ± SD RSI values were 0.36±0.17 (lung). 0.43± 0.18 (head and neck), 0.53±0.18 (rectal) and 0.12±0.22 (bladder). The difference in the mean RSI values between male and female patient cohorts was statistically significant in lung cancer only (p=0.0001). A large heterogeneity in GARD values were obtained across both female and male patient population, in all 4 cancer sites. The standard deviation of GARD values in male and female cohorts ranged from 0.51 (female head and neck) to 0.84 (male bladder). Mean GARD values were significantly different between the sexes in lung cancer only (p=0.0001).


Significantly higher GARD values in this cohort of female lung cancer patients supports reports that female sex is associated with a better outcome for this disease. This small in silico study suggests that further investigation of the impact of sex on radiation responses is warranted.