A trial assessing preoperative chemotherapy in patients with locally advanced but operable colon cancer

INTERVENTION: FOxTROT is a stratified, multi‐arm, multi‐site randomised trial platform for patients with locally advanced but operable colon cancer. All patients will have been assessed at a colon cancer MDT for review of pathology and radiology and potential participants will be identified. Patients meeting the registration eligibility criteria will be registered and a decision made about which FOxTROT comparison is most suitable. Participants will then be randomised in the appropriate comparison. FOxTROT 2 randomises participants between two trial arms: 1. Straight to surgery: patients will proceed straight to surgery as soon as possible and will be assessed for adjuvant chemotherapy as standard care. 2. Neoadjuvant chemotherapy followed by surgery. Participants in this arm will receive 6 weeks of OxFP chemotherapy with a choice (non‐randomised) between two regimens: OxMdG (2‐weekly oxaliplatin/calcium folinate/infusional fluorouracil) or OxCap (3‐weekly oxaliplatin with capecitabine). These should be delivered as per local practice. The clinician has two options of initial dosing as described above: full dose OxFp, or 80% dose OxFp. Treatment should start as soon as possible following randomisation. Patients should be reviewed prior to each cycle of treatment to assess for toxicity and any evidence of disease progression. Following completion of neoadjuvant chemotherapy, all participants will be reviewed in the oncology clinic to be assessed for adjuvant chemotherapy. Participants will be randomised using a central automated 24‐hour internet service based at the Leeds Clinical and Translational Research Unit (CTRU). CONDITION: Locally advanced but operable colon cancer ; Cancer ; Malignant neoplasm of colon PRIMARY OUTCOME: ; FOxTROT 2: Disease‐free survival (DFS), defined as the time from randomisation to disease recurrence, treatment failure, or death from any cause. The date of recurrence will be taken as the date of the CT scan which concluded disease recurrence. If a CT scan is not carried out, the date of recurrence will be taken as the date of the sample which indicated disease recurrence. Individuals who are lost to follow‐up or are alive and disease‐free at the time of analysis will be censored at their last date known to be alive and disease‐free.; FOxTROT 3: Tumour regression grade (TRG) (categorised as no response, mild, moderate, marked and complete response), measured at the time of surgery according to the modified Dworak grading system. DFS will be defined as per the FOxTROT 2 primary endpoint.; SECONDARY OUTCOME: ; 1. Tumour regression grade (TRG) measured according to the modified Dworak grading system at the time of surgery – FOxTROT 2 only; 2. Tumour regression score (TRS) (categorised as poor/no response, partial, near complete and complete response), measured at the time of surgery – FOxTROT 2&3; 3. Histopathological endpoints, measured from pathological samples at the time of surgery – FOxTROT 2&3:; 3.1. Tumour cell density; 3.2. Maximum tumour size; 3.3. Depth of invasion; 3.4. Apical node involvement; 3.5. Peritoneal involvement; 3.6. Nodal involvement; 3.7. R1/R2 resection rates; 4. Short‐term efficacy (and association with longer‐term outcomes) – FOxTROT 2 & 3:; 4.1. Downstaging by T‐stage, measured at pre‐registration, post‐NAC and 3‐years post‐randomisation; 4.2. Minimal residual disease by ctDNA and ctDNA alterations during neoadjuvant chemotherapy (NAC), measured from blood samples collected at baseline, prior to each cycle of NAC, post‐NAC and prior to adjuvant chemotherapy; 5. Safety and toxicity (both surgical and chemotherapy‐related) defined as the adverse reactions (ARs) and serious adverse events (SAEs) (including serious adverse reactions (SARs) and serious unexpected serious adverse reactions (?SUSARs)) reported on the trial according to CTCAE v5.0 and Clavien‐Dindo – FOxTROT 2 & 3; 6. Cancer‐related survival, defined as the time from randomisation to death caused by the same cancer, whether due to the original tumour or to a second primary same cancer. Individuals who are lost to follow‐up or are still alive at the time of analysis will be censored at their last date known to be alive. Death not related to cancer will be specified as a competing risk – FOxTROT 2 & 3; 7. Overall survival, defined as the time from randomisation to death from any cause. Individuals who are lost to follow‐up or are still alive at the time of analysis will be censored at their last date known to be alive – FOxTROT 2 & 3; 8. Surgical morbidity, defined as any surgery‐related complication within 30 days post‐surgery and surgical mortality, defined as death from any cause within 30 days post‐surgery – FOxTROT 2 & 3; 9. Patient‐reported outcomes (PROs) assessed according to outcomes measured on the EQ‐5D‐5L and QLQ‐C30 and CR29. PROs will be collected at baseline, post‐NAC, prior to adjuvant chemotherapy, 12 months post‐randomisation and 3 years post‐randomisation – FOxTROT 2 & 3; 10. Geriatric assessment scoring, will be determined by the collection of domains used to assess frailty, full details of which are provided in the FOxTROT 2 protocol. Geriatric assessment data will be collected at baseline, and prior to adjuvant chemotherapy – FOxTROT 2 only; INCLUSION CRITERIA: FOxTROT registration INCLUSION CRITERIA: 1. Biopsy‐confirmed adenocarcinoma of the colon (or upper rectum if too high for radiotherapy); high‐grade dysplasia is acceptable with unequivocal radiological evidence of invasive cancer* 2. Radiological stage T3‐4, N0‐2, M0 3. Patient being treated with curative intent 4. Tumour tissue is available for mismatch repair (MMR)/microsatellite instability (MSI) testing (local or central) 5. No clinical or radiological evidence of bowel obstruction 6. Age =18 years at the time of registration 7. Patient able and willing to provide written informed consent for the study * Patients with synchronous tumours are eligible if the most advanced tumour meets the criteria above (please note MMR/MSI testing requirements for randomisation depending upon the location of t