Molecular selection of therapy in metastatic colorectal cancer: the FOCUS4 molecularly stratified RCT

Background Complex trials with innovative designs are becoming increasingly common and offer the potential to improve patient outcomes in a shorter time frame. There is evidence that patients with colorectal cancer fall into different subgroups with varying responsiveness to therapy, and that this variation is linked to genetic biomarkers. To the best of our knowledge, FOCUS4 was the first molecularly stratified trial in metastatic colorectal cancer and remains one of the first umbrella trial designs to be launched globally. Objectives To identify novel therapies that improve disease control within the molecular subgroup of metastatic colorectal cancer in which the novel therapies were expected to be most effective. Design This was a Phase II/III molecularly stratified umbrella trial that used adaptive statistical methodology to decide which subtrial should close early; new subtrials were added as protocol amendments. Setting The maintenance setting following 16 weeks of first-line combination chemotherapy. Participants Patients with newly diagnosed metastatic colorectal cancer were registered, and central laboratory testing was used to stratify their tumour into molecular subtypes. Following 16 weeks of first-line therapy, patients with stable or responding disease were eligible for randomisation into either a molecularly stratified subtrial or the non-stratified FOCUS4-N trial. Interventions Of the 20 drug combinations that were explored for inclusion in the platform trial, three molecularly targeted subtrials were activated: FOCUS4-B (PIK3CA mutation or PTEN overexpression) – aspirin versus placebo; FOCUS4-C (TP53 and RAS mutation) – adavosertib (AstraZeneca Ltd, Cambridge, UK) versus active monitoring; and FOCUS4-D (BRAF-PIK3CA-RAS wild type) – AZD8931 versus placebo. A non-stratified subtrial was also carried out: FOCUS4-N – capecitabine versus active monitoring. Main outcome measures The main outcome measure was progression-free survival from the time of randomisation to progression, comparing the intervention with active monitoring/placebo. Toxicity and overall survival data were collected in all randomised patients, and quality of life (using EuroQol-5 Dimensions) data were collected in FOCUS4-N only. Results Between January 2014 and October 2020, 1434 patients were registered from 88 hospitals in the UK. Successful biomarker testing was completed in 1291 out of 1382 samples (93%), and 908 out of 1315 patients (69%) completing 16 weeks of first-line therapy were eligible for randomisation, with 361 randomly allocated to a subtrial. FOCUS4-B evaluated aspirin versus placebo in the PIK3CA-mutant/ PTEN -loss subgroup, but recruited only six patients, so was closed for futility. FOCUS4-C evaluated adavosertib versus active monitoring in 67 patients in the RAS + TP53 double-mutant subgroup and met its primary end point, showing an improvement in progression-free survival (median 3.61 vs. 1.87 months; hazard ratio 0.35, 95% confidence interval 0.18 to 0.68; p = 0022). FOCUS4-D evaluated AZD8931 in 32 patients in the BRAF-PIK3CA-RAS wild-type subgroup and showed no benefit, so was discontinued after the first interim analysis. FOCUS4-N evaluated capecitabine monotherapy versus active monitoring in 254 patients and met its primary end point, showing improvement in progression-free survival (hazard ratio 0.40, 95% confidence interval 0.21 to 0.75; p < 0.0001). Limitations FOCUS4-C and FOCUS4-N were closed early owing to COVID-19, so did not accrue their planned recruitment numbers. Conclusions Adaptive stratified medicine studies are feasible in common cancers but present challenges. Capecitabine monotherapy is an effective maintenance therapy. Wee1 inhibition using adavosertib shows significant clinical activity, notably in left-sided colorectal cancer. Trial registration This trial was registered as ISRCTN90061546. Funding This project was jointly funded by the Efficacy and Mechanism Evaluation (EME) programme, a MRC and National Institute for Health and Care Research (NIHR) partnership, and Cancer Research UK. This will be published in full in Efficacy and Mechanism Evaluation; Vol. 9, No. 9. See the NIHR Journals Library website for further project information.