Session Item

Saturday
November 28
10:30 - 11:30
Physics Stream 1
Proffered papers 5: Analysis for toxicity and outcome
1202
Proffered Papers
Physics
10:30 - 10:40
Peritumoural density as a biomarker of distant failure in NSCLC patients treated with SABR
OC-0096

Abstract

Peritumoural density as a biomarker of distant failure in NSCLC patients treated with SABR
Authors: Davey|, Angela(1)*[angela.davey@postgrad.manchester.ac.uk];van Herk|, Marcel(1);Faivre-Finn|, Corinne(1);Brown|, Sean(2);McWilliam|, Alan(1);
(1)The University of Manchester, Division of Cancer Sciences- Faculty of Biology- Medicine and Health, Manchester, United Kingdom;(2)The Christie Hospital, The Christie Hospital NHS Foundation Trust, Manchester, United Kingdom;
Show Affiliations
Purpose or Objective

Predictors of distant failure (DF) have not been established for NSCLC patients treated with SABR. Image markers have been proposed describing potential risk of microscopic disease extension (MDE), stratifying patients with loco-regional failure on dose to surrounding tissue (Salguero, 2013). In this work, we propose peritumoural density to describe MDE and risk of metastatic disease. We investigate peritumoural density taking dose variability into account.

Material and Methods

An in-house validated technique generated the GTV on the 50% 4DCT phase for 259 patients treated with SABR in 3, 5 or 8 fractions. Dose was converted to EQD2 with α/β=10. Local rigid registration was used to match tumour in each phase to 50% and calculate respiratory motion. Dose distribution was blurred according to respiratory motion to estimate delivered dose.

In-house data-mining was used to sample CT pixel values (inside lung) and dose at radial distance from the GTV. Briefly, we calculated the signed distance transform and created 2D cross-histograms of pixel value/dose versus distance for each patient. For every 1mm, dose standard deviation (SD) and mean pixel value was sampled (Fig.1).



Average cross-histograms of mean pixel value were calculated for patients who did and did not experience DF at 18 months, censored for follow-up. T-statistic describing difference in cross-histograms defined an important region, using permutation testing for statistical significance. Mean pixel value was sampled from this region.


Dose variability (SD) was sampled from the region (Fig.1A) extending outside the PTV, but within the expected MDE range (5-15mm), where high SD represents higher chance of underdosing MDE.

Patients were split into low and high dose variability based on the median. For each cohort, mean pixel value was assessed in univariable and multivariable Cox regression.
Results

The region of greatest density difference was identified as -3mm to 7mm (Fig.1B). On univariable analysis, there was no association with DF for mean pixel value or dose SD alone. Median dose SD was 8.4Gy.


However, higher pixel value predicts DF for high dose variability (HR=1.76, p=0.014), but not when dose is uniform (HR=1.12, p=0.665). This is true for continuous pixel value and split on the median (Fig.2A). HR for continuous pixel value represents increase in hazard per 100HU. This association remained in multivariable analysis correcting for tumour volume (Fig.2B). Mean pixel value and dose SD do not correlate with clinical variables.
Conclusion

High peritumoural density is prognostic for DF when surrounding dose is not uniform. Peritumoural density correlates with published biomarker GTV surface density, however, a larger region is sampled. The larger region may be less sensitive to observer variation, and capture information on tumour invasiveness which is likely important for risk of metastatic disease. Tumours with a high peritumoural density could benefit from increased margins. We aim to validate this work in an external cohort.