Automated segmentation and dose-volume analysis with DICOMautomaton

Incorporating parotid gland inhomogeneity into head-and-neck treatment optimization through the use of artificial base plans

Despite a great improvement in target volume dose conformality made possible in recent years by modulated therapies, xerostomia remains a common and severe side effect for head-and-neck radiotherapy patients. It is known that parotid glands exhibit a spatially varying dose response; however, the relative importance of subregions throughout the entire gland has yet to be incorporated into treatment plan optimization, with the current standard being to minimize the mean dose to whole parotid glands. The relative importance of regions within contralateral parotid glands has been recently quantified, creating an opportunity for the development of a method for including this data in plan optimization. We present a universal and straightforward approach for imposing varying sub-parotid gland dose constraints during inverse treatment planning by using patient-specific artificial base plans to penalize dose deposited in sensitive regions. In this work, the proposed method of optimization is demonstrated to reduce dose to regions of high relative importance throughout contralateral parotids and improve predictions for stimulated saliva output at 1-year post-radiotherapy. This method may also be applied to impose varying dose constraints to other organs-at-risk for which regional importance data exists.

A Practical Radiotherapy Treatment Planning Technique for Second-Incidence Cancers That Incorporates Complete Organ-At-Risk Dose History

INTRODUCTION

Patients requiring treatment for second cancer incidences present unique radiotherapy plan development challenges. Historical dose delivered to organs at risk must be accounted for to properly estimate lifetime toxicity risks, but historical dose delivered to the region now occupied by tumours does not contribute to the prescription dose. Treatment planning systems permit inclusion of a base plan but do not provide the ability to manipulate it. We present a technique, dose cropping, which incorporates organ-at-risk dose history into the base plan while selectively excising dose history to diseased tissues now occupied by tumours. A retrospective plan comparison is performed to assess the effectiveness of dose cropping.

METHODS AND MATERIALS

Nine patients who received a second course of radiotherapy for cancers of the head-and-neck were replanned using the proposed technique. Clinical second courses and replans were compared on the basis of conformity index, heterogeneity index, maximum point dose, tissue control probability (TCP), normal tissue complication probability (NTCP), and whether the planning guidelines could be met by the treatment planning system. Replan constraints and guidelines followed the clinical treatment. In addition, a tissue recovery model was incorporated, applied to both clinical and replan courses, and compared to estimate the relevance of the dose cropping technique in such regimes.

RESULTS AND DISCUSSION

Replans had reduced organ-at-risk maximum point doses (5 Gy for spinal cord, 4 Gy for brainstem), NTCP (2.9% median reduction), and were able to more consistently achieve the V_95% > 98% coverage target regardless of the tissue recovery model. At the same time, replans using the dose cropping technique were statistically indistinguishable from clinical second courses on the basis of plan conformity, heterogeneity, or TCP (P > .31 in all cases).

CONCLUSIONS

Dosimetric history cropping is a valuable and widely applicable technique for second cancer radiotherapy planning. It also provides a natural means to incorporate tissue recovery models, biologically effective dose conversion, and NTCP and TCP model evaluation.

Regional radiation dose susceptibility within the parotid gland: effects on salivary loss and recovery

PURPOSE

Xerostomia is one of the most likely late toxic effects of radiotherapy treatment in patients with head-and-neck cancers. Modern treatment techniques can incorporate knowledge of complication risk into treatment plans. To this end, the authors attempt to quantify the regional radiotherapy dose-dependence of salivary output loss and recovery in a prospective study.

METHODS

Salivary output was collected from patients undergoing radiotherapy treatment for head-and-neck cancers at the BC Cancer Agency between February 2008 and May 2013. Regional dose-dependence (i.e., dose susceptibility) of loss and recovery is quantified using nonparametric (Spearman's rank correlation coefficients, local linear regression) and parametric (least-sum of squares, least-median of squares) techniques.

RESULTS

Salivary flow recovery was seen in 79 of 102 patients considered (p < 0.0001, Wilcoxon sign rank test). Output loss was strongly correlated with left- and right parotid combined dose φ = min (DL,  45 Gy) + min (DR,  45 Gy), and can be accurately predicted. Median early loss (three months) was 72% of baseline, while median overall loss (1 yr) was 56% of baseline. Fitting an exponential model to whole parotid yields dose sensitivities A3m = 0.0604 Gy(-1) and A1y = 0.0379 Gy(-1). Recovery was not significantly associated with dose. Hints of lateral organ sub-segment dose-response dimorphism were observed.

CONCLUSIONS

Sub-segmentation appears to predict neither loss nor recovery with any greater precision than whole parotid mean dose, though it is not any worse. Sparing the parotid to a combined dose φ of <50 Gy is recommended for a patient to keep ≈40% of baseline function and thus avoid severe xerostomia at 12 months post-treatment. It seems unlikely that a population's mean recovery will exceed 20%-30% of baseline output at 1 yr after radiotherapy treatment using current (whole-organ based) clinical guidelines.