Background The combination of transarterial chemoembolization (TACE) with immune checkpoint inhibitors (ICIs) has emerged as a promising strategy for patients with unresectable hepatocellular carcinoma (uHCC). However, the comparative efficacy and safety of these regimens remain uncertain. We carried out a systematic review and network meta-analysis (NMA) of phase III trials evaluating TACE combined with immunotherapy-based treatments. Materials and methods PubMed/MEDLINE, Scopus, the Cochrane Library, and major oncology congresses were systematically searched for phase III trials comparing TACE alone with TACE plus ICI-based regimens. The primary endpoint was progression-free survival (PFS). Results Five phase III randomly assigned trials involving 2414 patients were included. Compared with TACE alone, TACE plus ICI-based strategies significantly improved PFS [hazard ratio (HR) 0.69, 95% confidence interval (CI) 0.63-0.77; fixed-effect model] and objective response rate [risk ratio (RR) 1.41, 95% CI 1.34-1.49]. Pooled overall survival estimate favored combination therapy but did not reach statistical significance (HR 0.88, 95% CI 0.78-1.00). Combination therapy increased grade 3-4 treatment-related adverse events (RR 2.49), serious adverse events (RR 3.10), and treatment-emergent adverse events leading to death (RR 1.73). Subgroup analyses seemed to suggest that patients with albumin-bilirubin (ALBI) score 1 could derive greater PFS benefit than those with an ALBI score of 2 or higher (interaction P = 0.0003). In the NMA, all regimens improved PFS versus TACE, and the P-score ranking for PFS identified atezolizumab plus bevacizumab plus TACE as the highest-ranked regimen (P-score 0.744). Conclusions Immunotherapy-based treatments combined with TACE consistently improved disease control in uHCC, although the survival benefit appeared more modest. Additional follow-up and head-to-head randomly assigned trials are needed to define the optimal systemic partner for TACE.
Background The incidence of early-onset colorectal cancer (eoCRC, <50 years) is rising, whereas late-onset CRC (loCRC, ≥70 years) is increasingly prevalent due to demographic shifts. CRC commonly metastasizes to the liver, impacting progression-free survival (PFS) and overall survival (OS). Current guidelines are mainly based on patients aged 50-70 years, and little is known about treatment and outcomes in younger and older patients with colorectal liver metastases (CRLM).Therefore, we compared treatment strategies and outcomes between eoCRC and loCRC patients with CRLM. Patients and methods Patients receiving their first local treatment of CRLM between 2018 and 2023 at Amsterdam UMC were extracted from the AmCORE registry and subdivided into three age groups: <50, 50-69, and ≥70 years. Primary and secondary endpoints were PFS and OS, respectively. Firth logistic regression evaluated factors influencing treatment choice. Results In total, 351 patients were included: 31 aged <50, 164 aged 50-69, and 156 aged ≥70. Younger patients received significantly more induction chemotherapy before first local treatment than those aged ≥70 and 50-69 (P < 0.001) and presented more frequently with a higher number of CRLM (P = 0.026). PFS was significantly shorter in patients <50 (P = 0.019), whereas OS did not differ significantly between age groups (P = 0.22). A larger number and size of CRLM were associated with an increased likelihood of induction systemic treatment. Conclusions Despite receiving more intensive treatment, younger patients did not show better outcomes: PFS was poorer and OS was similar compared with older patients. This may be explained by greater disease burden at presentation rather than intrinsically aggressive tumor biology.
Background Metastatic colorectal cancer (mCRC) shows organ-specific patterns of dissemination. Approximately 20% of patients develop disease confined to the liver throughout the course. Early recognition and characterisation of this phenotype may inform therapeutic stratification and improve outcomes. Patients and methods We retrospectively analysed patients with unresectable mCRC. Patients were classified by metastatic distribution into liver-limited, liver plus extrahepatic, or extrahepatic-only disease. Clinicopathological features, genomic alterations assessed by next-generation sequencing, and immune contexture based on a 12-gene signature from the primary tumour were evaluated. Results Among 1026 eligible patients, 20% had permanently liver-limited disease, 51% had both hepatic and extrahepatic involvement, and 29% had extrahepatic-only metastases. Liver-limited cases were enriched for synchronous presentation and nonrectal primaries, with a lower prevalence of KRAS mutations (P = 0.013). Tumours with liver plus extrahepatic dissemination showed greater genetic alterations across multiple cancer hallmarks (P < 0.05). Immune profiling revealed a markedly lower proportion of cold phenotypes in extrahepatic-only disease compared with liver-involved cases (P = 0.041). Despite similar genomic profiles, extrahepatic-only disease was associated with improved overall survival. Conclusions Permanent liver-limited mCRC defines a clinically and biologically distinct subgroup characterised by synchronous onset, nonrectal primaries, and reduced KRAS mutation frequency. The immune landscape of the primary tumour was associated with metastatic organ pattern.
Immunotherapy has revolutionized the treatment landscape of colorectal cancer (CRC), improving clinical outcomes in microsatellite instability-high tumors. In contrast, microsatellite stable (MSS) CRC exhibits intrinsic resistance to immune checkpoint inhibitors (ICIs), driven by reduced neoantigen load, resulting in an immunologically ‘cold’ tumor microenvironment (TME). To overcome these barriers, combination strategies integrating ICIs with tyrosine kinase inhibitors (TKIs) have been investigated based on their ability to inhibit multiple oncogenic and angiogenic signaling pathways. Beyond their antiangiogenic effects, preclinical evidence has demonstrated that TKIs can induce immunological reprogramming of the TME, promoting proinflammatory macrophage polarization, antigen presentation, vascular normalization, and T-cell infiltration. This evidence supported several early-phase clinical trials evaluating TKI-ICI combinations, showing encouraging signals of clinical activity in refractory MSS CRC. Recently, the interim analysis of the phase III STELLAR-303 trial demonstrated an overall survival benefit with the combination of zanzalintinib plus atezolizumab compared with regorafenib in pretreated MSS CRC. Notably, a consistent exploratory finding across the TKI-ICI trials was the potential role of liver metastases in driving local and systemic immunosuppression. Preclinical MSS CRC mouse models confirmed this suggestion, highlighting the association between the presence of liver metastases and impaired effector T-cell responses and antigen presentation. Overall, the heterogeneity of clinical outcomes underscores the complexity of immune resistance in MSS CRC and supports the development of multimodal biomarkers integrating spatial pathology, immune profiling, and artificial intelligence. This review discusses current preclinical and clinical evidence on TKI-ICI combinations in MSS CRC, exploring emerging mechanisms of immune resistance and potential therapeutic vulnerabilities.
Background Gemcitabine plus platinum combined with an immune checkpoint inhibitor (GP-ICI) is standard first-line therapy for advanced biliary tract cancer (BTC); however, tumor-site distribution, molecular epidemiology, and postprogression treatment access vary globally. We compared real-world outcomes and postprogression treatment patterns between US and South Korean tertiary centers. Patients and methods Patients with advanced BTC receiving first-line GP-ICI at MD Anderson Cancer Center (MDACC) or Asan Medical Center (AMC) between 2021 and 2025 were retrospectively analyzed. Outcomes included objective response rate (ORR), progression-free survival (PFS), overall survival (OS), and postprogression survival (PPS). Molecular profiling and treatment patterns were evaluated. Results Among 375 patients (MDACC, n = 121; AMC, n = 254), intrahepatic cholangiocarcinoma was observed at a higher proportion at MDACC (71.9% versus 42.1%), whereas extrahepatic cholangiocarcinoma was more common at AMC (10.7% versus 41.7%). ORR (21.5% versus 15.7%), median PFS (6.6 versus 7.2 months), and median OS (19.5 versus 14.9 months) were comparable. Median PPS was significantly longer at MDACC (13.1 versus 5.8 months; P = 0.002). IDH1 mutations and FGFR2 fusions were more common at MDACC, whereas HER2 alterations were more frequent at AMC. Among patients with actionable alterations, matched targeted therapy was administered more frequently at MDACC (75.8% versus 12.8%; P < 0.001) and was associated with significantly longer PPS [hazard ratio (HR) 0.15, 95% confidence interval (CI) 0.06-0.41, P < 0.001] and a trend toward longer OS (HR 0.54, 95% CI 0.26-1.10, P = 0.089). Conclusions Real-world GP-ICI outcomes were comparable between US and South Korean cohorts. The differences in molecular profiling, matched targeted therapy utilization, and postprogression outcomes underscore the need for broader molecular profiling and improved targeted therapy access in advanced BTC.
