ESTRO 37 Abstract book

S73

ESTRO 37

lead to the limit of ‘one-stop curation’. However, care must be taken that such approaches will not lead to more frequent and severe adverse effects. The steps already taken on this road, as well as the remaining hurdles will be discussed. SP-0147 Preserving bladder function with brachytherapy for muscle-invasive bladder carcinoma B. Pieters 1 1 Academic Medical Center, Radiation Oncology, Amsterdam, The Netherlands Abstract text Surgical therapy for muscle-invasive bladder cancer (MIBC) involves radical cystectomy resulting in loss of bladder function. In selected cases with limited tumor invasion in superficial muscle layers a trans urethral resection (TURB) can be considered. However, about half of the patients will experience a local relapse with a disease-specific mortality of 47%. Alternative bladder sparing therapies were developed with the great advantage that bladder function will be preserved. Standard treatments are trimodality therapies either as TURB, external beam radiotherapy (EBRT) and concomitant chemotherapy or TURB, EBRT, and brachytherapy. Selection to be treated with brachytherapy is a solitary tumor, less than 5 cm, T2 (limited T3) and not located in the bladder neck. Patients undergoing bladder conservative therapies would likely being happy because of retaining their bladder function. However, these patients will continue to be at risk for developing bladder recurrences, either non-muscle- invasive or muscle-invasive. Local recurrence rate is 20%. Of these recurrences about 40-50% are non-muscle- invasive for which can be treated with repeated conservative therapy (TURB with or without bladder instillations). Reported bladder function preservation after brachytherapy is 88%. About 90% of these patients have an unaltered bladder capacity. In conclusion high local control rate can be achieved in selected cases with brachytherapy. Bladder function is preserved in the majority of cases.

associated with resistance to chemoradiotherapy, increased propensity for metastases, and poor prognosis. Gene expression signatures have been devised to identify hypoxic tumours but the spatial heterogeneity in gene expression within a tumor raises the concern that multiple biopsies may be necessary to correctly identify the global tumor hypoxia status. The objective of this study was to determine if a genetic hypoxia signature is tolerant of intratumoral heterogeneity and whether a single biopsy may be sufficient to obtain a reliable Multiple biopsies (33) were obtained from 11 locally advanced cervical cancer patients prior to chemoradiotherapy. Between two and five different areas of each tumor were sampled. RNA was extracted and hybridized onto the Affymetrix U133 Plus 2.0 oligonucleotide chip. 5 validated hypoxia gene signatures were analyzed: 3 derived from clinical samples from various tumor sites (Winter, Buffa, Eustace) and 2 cervix- cancer specific signatures derived from DCE-MRI (Halle, Fjeldbo). The variance of gene expression within a patient, between patients, and the total variance were calculated. The ratio of between versus total variance (B/T) was calculated for each gene signature; the greater the B/T ratio, the greater the variance between tumors. The Spearman correlation coefficient was used to correlate gene expression signatures to each other and to compare signatures calculated using one biopsy and all the available biopsies. Results Patient and tumor characteristics were representative of typical cervix cancer patients treated with chemoradiotherapy: median age 47 (31-70), FIGO stage IB to IVB, and squamous histology (82%). The 5 hypoxia signatures analyzed were comprised of 6 to 99 genes each. The B/T ratio ranged from 0 to 1 for individual genes. The B/T ratios for the gene expression signatures were as follows: 0.79 (Fjeldbo), 0.71 (Halle), 0.69 (Buffa), 0.60 (Eustace), and 0.59 (Winter). Despite being derived in different ways and from different tumor types, the hypoxia gene expression signatures highly correlated with each other, with Spearman correlation coefficients ranging from 0.733 to 0.836. The Spearman coefficients between the hypoxia signature calculated using one biopsy and the same signature calculated using all available biopsies from the same tumor ranged from 0.855 to 0.891. Conclusion Compared with individual genes, hypoxia gene expression signatures are more consistent across multiple biopsies from different areas of an individual tumor and more tolerant of intratumoral heterogeneity. Cervix cancer- specific signatures had the best discriminatory ability in our patient population. In the setting where only a single biopsy is available, a cervix cancer-specific hypoxia gene signature may give a reasonable estimate of global hypoxia level and differentiate cervix cancers based on hypoxia. OC-0150 Assessing the immune contexture of anal squamous cell carcinoma D. Martin 1 , P. Balermpas 1 , R. Winkelmann 2 , U. Wieland 3 , M. Rave-Fränk 4 , C. Helmke 5 , M. Raab 5 , J. Kitz 6 , Y. Matthess 5 , K. Strebhardt 5 , C. Rödel 1 , F. Rödel 1 , E. Fokas 1 1 Klinikum der Johann Wolfgang Goethe Univ, Department of Radiotherapy and Oncology, Frankfurt, Germany 2 Klinikum der Johann Wolfgang Goethe Univ, Senckenberg Institute for Pathology, Frankfurt, Germany 3 University of Cologne, Institute of Virology- National Reference Centre for Papilloma- and Polyomaviruses, Cologne, Germany 4 University Medical Center Göttingen, Department of Radiotherapy and Radiation Oncology, Göttingen, Germany estimate of hypoxia. Material and Methods

SP-0148 Determinants of toxicity in pelvic radiotherapy R. De Crevoisier Centre Eugene Marquis, Rennes, France

Abstract not received

Proffered Papers: RB 2: Biomarkers

OC-0149 Intratumoral heterogeneity and hypoxia gene expression signatures in cervix cancer. J. Lukovic 1 , K. Han 1,2,3 , M. Pintilie 4 , N. Chaudary 5 , R. Hill 1,2,5,6 , A. Fyles 1,2,3 , M. Milosevic 1,2,3 1 Princess Margaret Cancer Centre, Radiation Medicine Program, Toronto, Canada 2 University of Toronto, Department of Radiation Oncology, Toronto, Canada 3 University of Toronto, Institute of Medical Sciences, Toronto, Canada 4 Princess Margaret Cancer Centre, Department of Biostatistics, Toronto, Canada 5 Princess Margaret Research Institute, Princess Margaret Cancer Centre, Toronto, Canada 6 University of Toronto, Department of Medical Biophysics, Toronto, Canada Purpose or Objective In response to hypoxia, cervical cancers modulate gene expression resulting in an aggressive phenotype that is