Background Colorectal cancer remains a leading cause of global cancer mortality, with metastatic CRC (mCRC) requiring sequential therapies after first line treatment failure. While regorafenib and fruquintinib are guideline-endorsed third-line options, their comparative value remains unestablished due to absent head-to-head trials. This real-world study evaluates clinical outcomes, safety, and cost differentials to model value-equilibrium pricing.Methods A retrospective cohort analysis included 25 mCRC patients (regorafenib: n = 5; fruquintinib: n = 20) treated at Sichuan Cancer Hospital (2021-2022) with follow-up through June 2023. Outcomes included real-world disease control rate (rwDCR), adverse events (CTCAE v4.03-graded), and daily treatment costs (medication, dose adjustments, adverse event management). A Monte Carlo simulation modeled cost equilibrium using Generalized Beta Distribution-derived adverse event variability.Results Baseline characteristics were balanced (median age: 58-63; 60%-70% male). rwDCR showed no significant difference (20% vs 25%, p = 1.000). Regorafenib demonstrated higher grade 3-4 toxicities (60.0% vs 20.0%), including hepatotoxicity (40.0% vs 15.0%) and hand-foot skin reaction (20.0% vs 0%). Fruquintinib exhibited unique hypertension (10.0%) and proteinuria (20.0%). Regorafenib incurred 75% higher daily costs (& YEN;455.53 vs & YEN;259.96, p = 0.001), primarily from medication expenses (& YEN;439.82 vs & YEN;253.71, p = 0.014). Pharmacoeconomic modeling identified regorafenib's value-based pricing threshold at 47.35% of current costs (& YEN;248.03/day; 95% CI: 247.98-248.09), revealing a 111% price-to-value mismatch.Conclusion Fruquintinib demonstrates comparable efficacy with superior safety and cost-effectiveness in third-line mCRC. Regorafenib's pricing exceeds its clinical value by twofold, underscoring systemic misalignment between drug costs and therapeutic benefit. These findings advocate for value-driven pricing reforms integrating toxicity-related economic burdens and provide a replicable framework for indirect treatment comparisons in oncology. However, the small sample size reduced statistical power, potentially biasing the findings.