Session Item

IMRT with IGRT followed by brachytherapy and concurrent chemotherapy is usually the mainstay of treatment in locally advanced cancer cervix (LACC). IMRT plans are usually generated on snapshot pre-external radiotherapy (ERT) CECT scans taken 1-2 weeks before starting ERT. These IMRT plans are generally considered to be applicable throughout the entire course of ERT. However, the gross tumor volume (GTV) in LACC is expected to undergo dynamic regression during the usual 5 weeks of ERT, representing a sigmoid dose-response curve. The study aims to analyze the accuracy of pre-ERT IMRT plans on various dose-volume histogram (DVH) parameters with the transposed GTV on re-CECT taken at 3-4 weeks of ERT.

15 consecutive patients of LACC were included in this prospective study. IMRT plans were generated 1-2 weeks before starting ERT to deliver 50Gy in 25 fractions over 5 weeks to 98% of the planning target volume (PTV). The same IMRT plan was superimposed on a re-CECT taken at 3-4 weeks of ERT. GTV and DVH parameters - V₅₀, V₉₅, D₉₅, D₉₈, D₅₀, D₅, D₂, conformity index (CI), homogeneity index (HI) and concordance index (ConI) were assessed and compared between pre-ERT and re-CECT scans. Additionally the V₅₀ of bladder and rectum; V₄₅ and D_190cc of small bowel were also evaluated. Before IMRT delivery, biweekly positional accuracy of the patient setup was monitored using portal imaging.

A total of 159 shifts in 15 consecutive patients before delivery of IMRT, each in X, Y and Z axes were recorded. The mean±SD of these shifts in X, Y and Z axes were 0.09±0.5cm, 0.12±0.5cm and 0.08±0.3cm respectively. Compared to pre-ERT IMRT, significant reduction in GTVs at 3-4 weeks were observed (p<0.0001). This resulted in significant differences in various DVH parameters between those of the pre-ERT PTVs vs. PTVs on re-CECT at 3-4 weeks of ERT V₉₅(p<0.001), D₉₈(p=0.011), CI(p=0.002), HI(p=0.009) and ConI (p<0.001), V₅₀ bladder (p= 0.003) and V₄₅ of small bowel (p<0.001)(Table). The mean differences inV₅₀, D₂, D₅ and D₅₀ were insignificant.

The significant reduction of the GTV in LACC at 4-weeks of ERT results in a marked variation in the key DVH parameters when compared to the pre-IMRT plan. Continuing the same pre-ERT IMRT plan for the entire course of 5 weeks of ERT, would result in inadvertent hot or cold spots, irradiate higher normal tissue and thus defeat the very purpose of IMRT. Periodic assessment of the dynamic changes in GTV is thus mandatory by early incorporation of adaptive IMRT plans to optimize dose to the GTV. This would ensure adequate intended dose coverage to GTV that could finally translate into improved clinical outcomes (Fig). Thus, IMRT can not only precisely hit the target but could even precisely miss the target, if tumor regression dynamics are not timely supplemented by adaptive IMRT plans. Mere setup verifications and their corrections are grossly inadequate to ensure adequate GTV-PTV coverage as evident on snapshot pre-ERT IMRT plans in LACC.