ESTRO 2021 Abstract Book

S283

ESTRO 2021

Serafini 6 , A. Devecchi 6 , C. Bergamini 7 , F. Badenchini 8 , T. Rancati 1 , R. Valdagni 8,9,10 , E. Orlandi 9,2 1 Fondazione IRCCS Istituto Nazionale dei Tumori, Prostate Cancer Program, Milan, Italy; 2 Fondazione IRCCS Istituto Nazionale dei Tumori, Radiation Oncology 2, Milan, Italy; 3 Fondazione IRCCS Istituto Nazionale dei Tumori, Medical Physics, Milan, Italy; 4 Fondazione IRCCS Istituto Nazionale dei Tumori, Platform of Integrated Biology, Department of Applied Research and Technology Development, Milan, Italy; 5 Fondazione IRCCS Istituto Nazionale dei Tumori, Medical Oncology 3, Milan, Italy; 6 Fondazione IRCCS Istituto Nazionale dei Tumori, Platform of Integrated Biology, Department of Applied Research and Technology Development, Milan, Italy; 7 Fondazione IRCCS Istituto Nazionale dei Tumori, Medical Oncology 3, Milan, Italy; 8 Fondazione IRCCS Istituto Nazionale dei Tumori, Prostate Cancer Program, Milan, Italy; 9 Fondazione IRCCS Istituto Nazionale dei Tumori, Radiation Oncology 1, Milan, Italy; 10 Università degli Studi di Milano, Department of Oncology and Hemato-Oncology, Milan, Italy Purpose or Objective A mono-institutional prospective trial was set up in 2017 to investigate the role of salivary microbiota (MB) in impacting on acute toxicity in head and neck cancer (HNCa) patients (pts) treated with radiotherapy (RT). We here investigated: i) the effect of RT dose (dose to the oral cavity, OC) in the modulation of oral MB at the genus and species level ; ii) the association between oral mucositis (OM) and MB modulation. Materials and Methods The study was conducted on 114 HNCa patients enrolled consecutively and treated with conventional (70Gy) or moderately hypofractionated VMAT in radical ( 66-70Gy@2-2.2Gy/fr ) or postoperative ( 50.4-66Gy@1.8- 2.2Gy/fr ) settings with or without systemic therapy. For each patient, two salivary samples (before RT and within a week of RT end) were collected using OMNIgene ORAL-OM501 (DNAGenotek) devices. Microbial DNA was extracted using the QIAsymphony DSP Virus/Pathogen (Qiagen). Bacteria were identified by 16S rRNA gene sequencing, using the Ion 16S Metagenomics kit (Thermo Fisher Scientific) and pooled in Operational Taxonomic Units with Uclust software. The modulation at the RT end was evaluated by Wilcoxon’s signed-rank test for paired samples; multiple testing was adjusted using the Bonferroni correction. Analyses were performed on the whole population and after stratification of patients by the maximum grade of OM (G1 vs G2 vs G3, as for CTCAE v. 5.0) and quartiles of the mean dose to the OC. Results 77 genera and 225 bacterial species were identified in at least 10 % of the population. RT significantly modulated 26 genera and 61 species in the whole cohort (at p level < 0.01). The genera and species with modulation ≥2 orders of magnitude are shown in fig.1a and fig.1b. 12 (11 %), 62 (54 %), and 40 (35%) pts reported G1, G2 and G3 OM, respectively (no pts scored as G0). After stratification by OMs grade, 12 genera were significantly modulated by RT within G2 and 6 within G3 (fig. 1c); 18 species within G2 and 17 within G3 (fig.1d). No significant MB modulation was found within G1 OMs (at p level<0.01). Detailed results about MB significantly modulated by RT, which were found within both G2 and G3 OMs, are shown in fig.1c and fig.1d. The analysis including RT resulted in 3 species ( Fusobacterium periodonticum, Gemella sanguinis and Haemophilus parainfluenzae ) showing a dose-response behaviour, i.e., increased modulation with increasing OC mean dose. Particularly, OC mean dose > 38 Gy relevantly affected the abundance of Gemella sanguinis (fig.2b), which disappears almost entirely at higher doses.