INTEGRATE II: randomised phase III controlled trials of regorafenib containing regimens versus standard of care in refractory Advanced Gastro-Oesophageal Cancer (AGOC): a study by the Australasian Gastro-Intestinal Trials Group (AGITG)

The aim of INTEGRATE IIa is to determine whether regorafenib160mg daily on day 1 to 21 of a 28-day cycle improves OS compared to placebo. The aim of INTEGRATE IIb is to determine whether regorafenib 90 mg daily on day 1 to 21 of a 28-day cycle in combination with nivolumab 240 mg every 2 weeks improves OS compared to investigator’s choice of chemotherapy in participants with AGOC who have progressed or were intolerant to two or more lines of chemotherapies including platinum and fluoropyrimidine based regimens. After 2 months, nivolumab can be given monthly at a dose of 480 mg.

INTEGRATE IIa is a prospective, multi-centre, comparative, double-blinded phase III RCT where participants with histological or cytological confirmation of AGOC or undifferentiated carcinoma of primary gastro-oesophageal origin are randomised 2:1 to receive either regorafenib monotherapy (Arm A) or placebo (Arm B) (Fig. 1). Randomisation is stratified by location of tumour (GOJ (gastro-oesophageal junction) vs. gastric), geographic region (Asia vs. rest of world) and prior VEGF inhibitors (yes vs no).

INTEGRATE IIb is a prospective, multi-centre, comparative, open label phase III RCT where participants with histological or cytological confirmation of AGOC or undifferentiated carcinoma of primary gastro-oesophageal origin are randomised 2:1 to receive regorafenib and nivolumab (experimental arm) or investigator’s choice of chemotherapy (control arm) (Fig. 2). Randomisation is stratified by geographic region (Asia vs. rest of world), prior VEGF inhibitors (yes vs no) and prior immunotherapy (yes vs no).

For INTEGRATE IIa and INTEGRATE IIb, all treatments are to be administered until disease progression, unacceptable toxicity, withdrawal of consent, more than 2 dose reductions due to treatment related adverse events or day 1 treatment delays of longer than 28 days. Randomisation will be performed centrally. Participant quality of life (QOL) questionnaires are to be completed on cycle 1 day 1 of treatment and every 4 weeks thereafter during treatment and at the time of disease progression. Participants who end treatment before disease progression will have QOL data collected every 8 weeks until disease progression. A list of participating centres is provided in Table 1 for INTEGRATE IIa and Table 2 for INTEGRATE IIb.

The primary end point for INTEGRATE IIa and INTEGRATE IIb is overall survival in each arm, defined as the interval from date of randomisation to date of death from any cause or the date the patient was last known to be alive. Secondary objectives for INTEGRATE IIa and INTEGRATE IIb are PFS, ORR (according to RECIST v1.1), QOL and safety profile (rates of adverse events per CTCAE v4.03). The tertiary and correlative endpoints may include: investigation of VEGF-related biomarkers and biomarkers relating to angiogenesis and/or tumourigenesis in blood and tumour as prognostic and/or predictive markers for study endpoints relating to survival, response and safety; regorafenib levels to evaluate pharmacokinetics in participants from different geographical regions; evaluation of the prevalence and distribution of the four proposed molecular phenotypes of gastric cancer proposed by the Cancer Genome Atlas Research Network [17], and their association with angiogenic biomarkers and regorafenib activity. Additional correlative endpoints for INTEGRATE IIb may include associations between autoimmunity and clinical outcomes, immune profiling of tumour, tumour mutational burden and cellular and molecular signatures associated with immune related adverse events.

INTEGRATE IIa and INTEGRATE IIb have similar eligibility criteria:

Participants aged ≥ 18 years with ECOG performance status < 2, who have histological or cytological confirmation of metastatic or locally advanced gastro-oesophageal carcinoma (gastric or GOJ cancer). Participants must have failed or be intolerant to a minimum of 2 lines of prior anti-cancer therapy and must have had at least one platinum agent and one fluoropyrimidine analogue.

Other main eligibility criteria include:

Main exclusion criteria include:

In INTEGRATE IIa, participants enrolled in Arm A will receive regorafenib 160 mg (4 × 40 mg tablets) orally once daily on days 1 to 21 of a 28-day cycle until disease progression, unacceptable toxicities, day 1 treatment delay of more than 28 days or withdrawal of consent. Participants enrolled in Arm B will receive matching placebo. Missed or vomited tablets cannot be compensated for by treatment later.

In INTEGRATE IIb, participants enrolled in the experimental arm of the study will receive regorafenib and nivolumab. Regorafenib will be self-administered orally once daily at a dose of 90 mg (3 × 30 mg tablets) on days 1 to 21 of a 28-day cycle until disease progression, unacceptable toxicities, day 1 treatment delay of more than 28 days or withdrawal of consent. Only one dose level reduction of regorafenib is permitted (90 mg to 60 mg) for management of regorafenib related adverse events. If regorafenib is not tolerated at a dose of 60 mg, regorafenib will be discontinued. Nivolumab will be administered intravenously at a dose of 240 mg on day 1 of a 14-day cycle until disease progression or prohibitive adverse events as per protocol or up to 24 infusions or 2 years maximum treatment duration. After 2 months, nivolumab can be administered at a dose of 480 mg on day 1 of a 28-day cycle. If a delay in nivolumab administration is required when a treatment cycle has commenced, nivolumab cannot be delayed more than 12 days for the 14-day schedule, and not more than 20 days for the 28-day schedule. If nivolumab needs to be delayed more than 12 weeks from last treatment date due to toxicity, nivolumab will be discontinued permanently. The continuation of regorafenib monotherapy is permitted despite permanent discontinuation of nivolumab. Participants enrolled in the control arm will receive investigator’s choice of chemotherapy with either docetaxel 60-75 mg/m² intravenously every 21 days, or paclitaxel 135-250 mg/m² intravenously on day 1 every 21 days, or paclitaxel 80 mg/m² intravenously on days 1, 8, 15, 21 every 28 days, or irinotecan 250-300 mg/m² intravenously on day 1 every 21 days, or irinotecan 150-180 mg/m² intravenously on days 1 and 15 every 28 days, or irinotecan 125 mg/m² intravenously on days 1 and 8 every 28 days, or trifluridine and tipiracil 35 mg/m² up to 80 mg dose orally twice daily on days 1 to 5 and 8 to 12 of a 28 day cycle.

All participants in INTEGRATE IIa and INTEGRATE IIb will receive best supportive care which includes any intervention to preserve comfort and dignity such as stent insertion for gastric outlet obstruction, radiation to a bleeding primary tumour and use of any concomitant medications considered necessary for participant’s wellbeing including antiemetics, antidiarrhoeals, anti-inflammatory agents and analgesics.

Disease assessment by CT scan will be performed at baseline and every 8 weeks, regardless of delays in drug administration until disease progression. QOL questionnaire (European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 and EORTC QLQ-STO22) will be completed at baseline and every 4 weeks during treatment and at the time of disease progression. For participants who end treatment before disease progression, QOL questionnaire will be collected every 8 weeks during follow up until disease progression. Tumour marker CA19-9 is assessed at the same time as CT imaging until disease progression.

Blood for biomarker research is collected from all participants at four timepoints: Cycle 1 Day 1, Cycle 2 Day 1, Cycle 4 Day 1, and end of treatment. Blood for pharmacokinetics (PK) is collected at three timepoints: Cycle 1 Day 15, Cycle 2 Day 1, and Cycle 2 Day 15 (optional) drawn before regorafenib is taken and 1 to 4 h afterwards. All blood samples collected are processed, frozen and biobanked. Archival tissue from the primary tumour or metastases are retrieved for translational research.

The primary analysis for INTEGRATE IIa and IIb is a comparison of OS between the experimental and control arms using a log-rank test accounting for stratification factors. A HR for OS will be estimated from a stratified Cox proportional hazards model accounting for stratification factors. PFS will be analysed in a comparable fashion to OS. ORR will be compared using Cochran-Mantel–Haenszel test accounting for stratification factors or Fisher’s exact test if expected cell counts are low. Quality of life will be analysed using the approaches applied to the INTEGRATE trial [18]. A series of subgroup analyses will be performed on OS for the stratification factors used at randomisation.

When the INTEGRATE platform protocol was proposed, the original sample size target for INTEGRATE II (regorafenib vs placebo) of N = 350 was changed to N = 250 for the INTEGRATE IIa trial. A study with 250 participants randomised in a 2:1 ratio (regorafenib: placebo) to yield an expected 221 OS events and have 80% power to detect a HR for OS of 0.67 with a 2-sided α of 5%. The OS events from INTEGRATE IIa will also contribute to a pooled analysis with the previous INTEGRATE trial which had 123 OS events. The conditional power of this combined pooled analysis will have at least 90% power to detect a HR for OS of 0.67. To determine the appropriateness of a pooled analysis, heterogeneity of treatment effect on OS across INTEGRATE and INTEGRATE IIa will be tested. Assuming there is no statistically significant heterogeneity, a sequential closed testing procedure (with a 2-sided α of 5%) will be applied to the following null hypotheses of no treatment effect in the:

For INTEGRATE IIb, a sample of 450 participants will be randomised 2:1 (regorafenib and nivolumab: chemotherapy) and followed until 380 deaths occur to provide at least 90% power to detect a HR for OS of 0.70 with a 2-sided α of 0.05. The design also accommodates early stopping for benefit/harm at an interim analysis performed at 2/3 of the required events using the error spending approach of Lan-DeMets. A safety analysis will be undertaken when a minimum of 8 participants are enrolled from Japan, 16 from Asia and the remainder from USA, Europe and Australasia. The final analysis on OS will involve testing the following two null hypotheses: (1) no treatment effect on OS in the whole trial cohort (\({H}_{0}^{All}\)); and, (2) no treatment effect on OS in the Asian region cohort (\({H}_{0}^{Asian}\)). A sequenced closed testing procedure will be used to constrain the overall type I error of these two tests to 5%.

All adverse events for INTEGRATE IIa and INTEGRATE IIb will be recorded from the first dose of study treatment until 30 days (90 days for nivolumab) following last treatment dose. The investigator is responsible for ensuring all adverse events observed by the investigator or reported by the trial participants are documented in electronic case report forms (eCRFs). Serious adverse events (SAEs), including suspected unexpected serious adverse reactions (SUSAR) occurring during the study must be reported to the sponsor within 24 h of investigational site staff becoming aware of the event according to local procedures. The sponsor is responsible for the medical review of all SAEs and for their notification to the appropriate ethics committees and local authorities.