Risk of disease progression in first-line metastatic colorectal cancer therapy to guide disease reassessments―analysis of 11 trials by AIO and GONO

BACKGROUND: We evaluated the distribution and risk of disease progression (PD) in first-line therapy of unresected metastatic colorectal cancer (mCRC) patients receiving chemotherapy + biologics. The aim of the analysis is to provide guidance for the timing of disease reassessments during first-line therapy. PATIENTS AND METHODS: Individual data of 2939 unresected patients from TRIBE, MOMA, TRIBE2, VALENTINO, ATEZOTRIBE, TRIPLETE, FIRE-3, XELAVIRI, PANAMA, FIRE-4 and FIRE-4.5 were analyzed. The frequency and risk of PD events were calculated for individual timepoints during therapy. RAS/BRAF profiling, tumor sidedness, type of therapy and early tumor shrinkage (ETS) were used to identify subgroups for risk assessment. A Cox regression model to predict first-line progression-free survival (PFS) was built. RESULTS: In the overall population, the maximum frequency of PD events was observed at 7.6 months, with an absolute PD risk of 19%. Then, the PD risk flattened, achieving a maximum of 23% at 14 months in RAS/BRAF-wild-type patients (n = 1786), 25% at 10 months in RAS-mutant patients (n = 973) and 35% at 8 months in BRAF-mutant patients (n = 180). Eastern Cooperative Oncology Group performance status >0, right-sidedness, initially unresected primary tumor, higher number of organs involved by metastases and BRAF mutation were independently associated with a higher risk of PD in first line. The impact of baseline characteristics on PFS was mitigated after incorporation of ETS in the model. CONCLUSIONS: The distribution of PD events does not follow a Gaussian pattern, with the highest density observed between the third and fourth reassessment of a bimonthly surveillance schedule. In clearly unresectable patients, restaging should focus on the interval between 6 and 10 months and not on the initiation of systemic therapy. Our model might be helpful to schedule radiological reassessments according to baseline characteristics, early response and the expected duration of each treatment efficacy.