ESTRO 38 Abstract book

S426 ESTRO 38

This analysis suggests that CRC metastases have an overall poor response rate to stereotactic radiotherapy. However, it seems that SRS should prefer over HFSRT in patients with limited CRC brain metastases. Moreover, it demonstrates that BPFS directly impact OS. PO-0816 Adaptive radiotherapy concomitant with chemotherapy as preoperative treatment for rectal cancer P. Passoni 1 , N. Slim 1 , C. Fiorino 2 , C. Gumina 1 , M. Ronzoni 3 , V. Burgio 3 , A. Fasolo 3 , F. De Cobelli 4 , A. Palmisano 4 , U. Elmore 5 , P. De Nardi 5 , A.M. Tamburini 5 , A. Vignali 5 , R. Rosati 5 , R. Calandrino 2 , N. Di Muzio 1 1 Ospedale San Raffaele IRCCS, Radiation Oncology, Milan, Italy ; 2 Ospedale San Raffaele IRCCS, Medical Physics, Milan, Italy ; 3 Ospedale San Raffaele IRCCS, Medical Oncology, Milan, Italy ; 4 Ospedale San Raffaele IRCCS, Radiology, Milan, Italy ; 5 Ospedale San Raffaele IRCCS, Surgery, Milan, Italy Purpose or Objective To report our clinical experience with hypofractionated, adaptive radiotherapy (RT) concomitant with chemotherapy (ChT) as preoperative treatment for rectal cancer. Material and Methods Patients (pts) with T3/T4N0 or any TN+ rectal adenocarcinoma were included in an observational study. Chemotherapy consisted of Oxaliplatin 100mg/m 2 on days -14, 0, +14 and c.i. 5FU 200mg/m 2 /day or oral capecitabine 825mg/m 2 x2/day (from 2015 on) from day – 14 to the end of RT. RT started on day 0, was delivered by Tomotherapy, and was divided into two phases of 12 and 6 fractions (Frs), respectively. Pts were positioned on Comby-Fix ® and underwent a simulation CT and MR, subsequently matched. In the first RT phase CTV included mesorectum, perineum in case of tumor < 6cm from anal canal, and regional lymph-nodes. PTV was defined as CTV with a margin of 0.5 cm and received 27.6 Gy in 12 fractions (2.3 Gy/Fr). A simulation CT and MR were repeated after two cycles of ChT and 9 Frs of RT. Two new volumes were considered: GTVadaptive, defined as the residual tumour (T and N) visible on the intermediate MR images, and PTVadaptive, created from GTVadaptive with a margin of 0.5 cm. In the second RT phase, the adaptive phase, PTV again received 2.3 Gy/fr x 6 Frs (total dose 41.4 Gy in 18 Frs), while PTVadaptive received a simultaneous integrated boost of 3.1 Gy/Fr x 6 Frs (total dose 46.2 Gy in 18 Frs). Results From 9/2009 to 5/ 2017, 82 pts were prospectively enrolled, T2=5, T3=72, T4=5, N+=67. Median distance from anal canal was 6.5 cm (0 cm in 18 pts), median cranium- caudal tumor length was 5 cm (1.2-11 cm). No G4 toxicity occurred; G3 toxicity consisted of diarrhoea in 10/82 pts (12%) and proctitis in 4 pts (5%). Diarrhoea always occurred before the adaptive phase. Eighty pts (97%) completed RT; median duration of RT was 25 days. Fifty- seven (70%) and 74 (84%) received > 75% of oxaliplatin and 5FU dose, respectively. Two pts achieved complete clinical response and refused surgery, 1 pt was lost, 1 pt had early distant progression. Seventy-eight pts were resected (73 R0, 5 R1), 72 pts had conservative surgery, 6 pts had abdominal-perineal amputation. Twenty-two pts (28%) had pathological complete response, 23 pts (29%) had TRG3 with 1-5% of residual vital cells. Fifty-two pts received adjuvant chemotherapy. At a median follow up of 30 months, calculated from the end of chemoradiotherapy to the last clinical check or death, 3 pts had local and 20 pts distant relapse. Local-regional- relapse free survival, distant metastasis free survival, and overall survival were 96.5%, 75.7% and 95.7%, respectively.

BM criteria. A Bayesian multivariate logistic regression was performed to evaluate the influence of several factors on the probability of response. Secondary endpoints included overall survival (OS) and brain progression-free survival (BPFS). Results This international, multicentric retrospective analysis involved 58 patients and 69 metastases. The median response rate for metastasis treated was 65.9% (CI 95% [51.9% – 79.9%]) and was positively influenced by SRS whereas multiple brain metastases were related to poorest response rates. The results of the Bayesian multivariate logistic regression are reported in Table 1.

Odds Ratio

Standard deviation 2.5% 95% p(OR<1) p(OR>1)

Disease controlled 0.73 0.51

21%

0.17 2.05 79%

Multiple brain metastasis

0.72 0.52

0.16 2.05 80%

20%

PTV (mL) 0.99 0.03

0.94 1.04 70% 0.36 13.5 21%

30% 79%

SRS

3.27 4.12

The median follow-up from initial treatment was 31 months. The median overall survival (OS) from initial treatment was ten months (CI 95% [5 – 22]), the Kaplan- Meier survival curve is reported in Figure 1. OS was negatively influenced by extra-cranial metastases, male gender, whereas it influenced positively by brain progression-free survival (BPFS).

Median BPFS was not met. BPFS Kaplan Meier survival curve is reported in Figure 2.

Conclusion We report the results of one of the largest cohort of CRC brain metastasis treated with stereotactic radiotherapy.