Volumetric analysis of anal cancer relapses following radical chemoradiation.
576 Background: Definitive chemoradiotherapy is standard of care in anal squamous cell carcinoma. The ACT II trial set the standard achieving three year overall survival rates of 73%. However patients with locally advanced disease have a ~50% local relapse rate. Studies have failed to demonstrate an improvement in local relapse rate by altering the systemic therapy. Although 2/3 of relapses are local as opposed to regional or distant, there is limited knowledge on the 3D position of these relapses. We aim to retrospectively review the patterns of local failure in three dimensions, relative to the radiotherapy plan. Methods: Between February 2007 and April 2012, 77 patients were treated for squamous cell carcinoma of the anus with radical chemoradiotherapy. Early stage (T0/2) and locally advanced (T3/4) tumours constituted 53% and 44% respectively. 37% of patients had nodal disease. As per ACT II protocol we used a gross tumour volume (GTV) to PTV margin of 3cm and prescribed to a dose of 50.4Gy. In patients with a local recurrence, imaging from the time of recurrence was imported into the radiotherapy planning system. The volume and site of gross recurrence was noted relative to the previous GTV and 95% isodose line. The mean dose delivered to the relapsed area was calculated. Results: Median follow-up was 42 months (range 14 to 78 months). Eight (10%) patients failed to respond. Six (8%) patients developed an isolated local relapse, 2 (3%) distant metastatic disease alone and 1 both local and distant relapse. Of those that relapsed locally, 5 patients were T2N0, one T3N0 and 1 node positive. There were no regional relapses. The entire macroscopic relapse lay within the 95% isodose line of the 50.4Gy in all patients. The median percentage of relapse that lay within GTV was 63% (range 29 to 100%). A GTV to PTV margin of 2cm would have encompassed all macroscopic relapses. The median doses to relapsed area was 52Gy (range 51Gy to 53Gy) with the minimum dose to any relapsed area being 49Gy. Conclusions: The majority of relapses centred on the GTV suggesting the gross tumour was uncontrolled with current doses in these patients. A reduction of GTV to PTV margin may be appropriate, facilitating trials of dose escalation with the aim of improving outcomes.