Session Item

To evaluate quality assurance of preoperative radiotherapy in the CRITICS-II trial.

The CRITICS-II trial is enrolling patients (pts) with resectable, clinical stage IB-IIIC histologically proven adenocarcinoma of the stomach or gastro-esophageal junction (GEJ). Patients are randomized between (1) preoperative (preop) chemotherapy, (2) preop chemotherapy plus chemoradiotherapy (CRT) or (3) preop CRT. Chemotherapy consists of 4 cycles of docetaxel, oxaliplatin and capecitabine (DOC) in arm 1 and 2 cycles of DOC in arm 2. Chemoradiotherapy consists of 45 Gy in 25 fractions of 1.8 Gy plus weekly paclitaxel and carboplatin (Figure 1). A delineation atlas has been made available. For pts with a GEJ/cardia tumor lymph node stations (LNS) according to the Japanese classification system 1-4, 7, and 9-13 should be included in the clinical target volume (CTV). For pts with corpus tumor LNS 3-13 should be included in the CTV and for pts with an antrum tumor LNS 3-9; 11-13 should be included in the CTV. Delineations of participating centers are centrally reviewed by one of the two radiation oncologists of the coordinating center prior to start of CRT. Noncompliance was defined as too narrow delineation/missed target of stomach and/or LNS: minor if it applied to 1-2 CT-slices and major if it applied to ≥3 CT-slices. Contamination was defined as too wide delineation of the stomach and/or LNS outside the intended CTV delineation: minor if it applied to 1-2 CT-slices, major of it applied in ≥3 CT-slices.

From 2018, a total of 133 pts have been included in the CRITICS-II trial. Among them, 88 pts were intended to receive CRT. To date, 80 delineations have been reviewed. Three delineations are awaited and five patients did not start CRT treatment as planned. For 20 out of 80 patients (25%), suggestions for CTV modification were communicated and in the large majority (85%) these suggestions were completely implemented. In 2018, suggestions were made in 6/11 (55%) of delineations. In 2019, 2020 and 2021 these rates were 6/26 (23%), 4/22 (18%) and 4/21 (19%), respectively (p=0.139). Contamination did not occur. Major noncompliance occurred in 11 cases (14%) and minor noncompliance occurred in 7 cases (9%). Both minor and major noncompliance occurred in one case (1%). Type of noncompliance was unknown in one case. Major noncompliance included: missed target/too narrow delineation of LNS in eight cases (10%) and both in four cases (5%). Minor noncompliance included: too narrow delineation of the stomach in one case (1%), missed target/too narrow delineation of LNS in five cases (6%) and both in two cases (3%). More details on noncompliance of LNS is shown in Table 1.

Systematic quality assurance of target delineations is an important instrument to evaluate and improve radiotherapy quality. In a sequential cohort of patients receiving preoperative therapy in the CRITICS-II trial, adaptations for the delineation had been suggested in 25%. Adaptation rate decreased over time.