Abstract

Outcomes in locally recurrent rectal cancer treated with reirradiation stereotactic ablative radiotherapy (SABR)

A diagnosis of locally recurrent rectal cancer (LRRC) is associated with considerable morbidity and a poor prognosis. While surgical series demonstrate a median survival of 37 months; the median survival of non-operative management is approximately 10 months. The use of re-irradiation stereotactic ablative body radiotherapy (SABR), offers an alternative to extensive surgery. However designing trials in this setting is challenging due to the limited available data. We performed a multicentre, retrospective case series of patients treated with SABR reirradiation, with a view to identifying appropriate outcome measures in clinical trials.

All patients in three centres treated with 30Gy in 5 fractions reirradiation were identified through local radiotherapy systems. Patients were treated with 30Gy in 5 fractions prescribed to D95. Both gantry based linear accelerators and Cyberknife were acceptable platforms for treatment delivery. Follow up consisted of review with CT scans at 3, 6, 12, 18 and 24 months. Radiotherapy systems provided details of treated lesions, appropriate local hospitals and general practitioners were contacted for information regarding site of relapse, date of relapse, details on last follow up and death. Statistical analysis was carried out using R version 3.6.1 (R Core Team (2019). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/)

70 patients received reirradiation SABR for rectal cancer between the 3rd August 2015 and 2nd September 2020. 8 patients were excluded, 7 received SABR for an R1 resection, 1 patient had lung metastasis at outset; 62 patients remained with 73 lesions. Using the reverse Kaplan-Meier method, median FU was 31 months. Median age was 65 years (range 36 to 86). Male to female ratio 60/40. 59 patients had lymph node recurrences, 1 bony recurrence, 1 penile and 1 perianastomotic recurrence. Median GTV volume of the largest lesion in each patient was 13.4cm3 (1 to 122cm3). Other than survival data, results are based on 61 patients as one patient moved out of area and only date of death is available. Median (95% CI) PFS was 12.1months (10.4, 16.6) with 2 year PFS rate of 0.23 (0.14, 0.40). 42 of 61 (69%) had progressed, initial sites of progression were local, distant and both local and distant in 18, 11 and 12 patients. 23 (37%) had died, 14 (64% of deaths) with consequences of uncontrolled pelvic disease contributing to death; for example tumour related fistulas, irretractable pain, hydronephrosis and issues associated with stents, bowel obstruction and recurrent PR bleeds. 28 of 61 (48%) had a local recurrence at the time of death or last follow up. Of 39 patients alive at FU, 14 (36%) were living with a local relapse. Median OS was 38.7months (28.8 -), 2 year OS rate 0.76 (0.64, 0.90).

Historic median OS in non-operative management of LRRC is approximately 10 months. The high rate of eventual local recurrence, percentage of deaths due to consequences of local disease and the encouraging median OS of 38.7 months in our series; suggests dose escalation using a SABR technique may be beneficial and OS may be an appropriate primary endpoint for a trial. Given the high percentage of patients living with a local recurrence at last follow up, with all the associated morbidity, quality of life must form a major part of future trials.