Biliary tract cancers: topaz-1 up-date, a practice changing study

Biliary tract cancers (BTCs) are an increasing public health concern due to their rising incidence—particularly among individuals under 60 years of age—their frequent late-stage diagnosis, and their poor prognosis, with a 5-year survival rate of less than 20% and a median overall survival (OS) of around 12 months (1,2). BTCs originate from the intrahepatic or extrahepatic bile ducts or the gallbladder. This anatomical heterogeneity is compounded by diverse risk factors, resulting in a complex disease landscape at the molecular level. Over the past decade, significant progress has been made in unraveling the molecular characteristics of these neoplasms, enabling advances in selected tumor subgroups through the development of targeted therapies and personalized medicine approaches (e.g., FGFR2 fusions, IDH1 mutations, HER2 amplification or overexpression, BRAF mutations) (3,4). Despite these advances in molecularly defined subsets, the vast majority of patients with BTC still face dismal outcomes, underscoring the urgent need for effective systemic strategies accessible to all patients, not only to biomarker-selected minorities.

Immunotherapy has also found a role in the management of advanced, non-curable BTCs (aBTCs) as a first-line treatment, in combination with the standard chemotherapy regimen of gemcitabine and cisplatin (GC), which has been the reference since the ABC-02 trial (5). The TOPAZ-1 trial was a randomized, double-blind, placebo-controlled phase 3 study conducted across 17 countries in Asia, Europe, and the Americas (6). It was the first real test of immunotherapy as a frontline partner to chemotherapy in this disease. Eligible patients had previously untreated aBTCs (or relapsed aBTCs with completion of adjuvant therapy more than 6 months earlier). Carcinomas of the ampulla were excluded. All patients had an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. A total of 685 patients were randomized 1:1 to receive either durvalumab (1,500 mg) or placebo in combination with GC as first-line therapy. The anti-PD-L1 immunotherapy was administered on day 1 of each 3-week cycle. After eight cycles of GC, durvalumab was continued as monthly maintenance therapy.

In 2022, TOPAZ-1 met its primary endpoint in a preplanned interim analysis, demonstrating a significant improvement in OS with the addition of durvalumab to GC compared to placebo with GC in patients with aBTCs (6). While the absolute gain in median survival was modest—barely over 1 month—the durability of benefit seen in a fraction of patients differentiates this study from prior negative efforts and explains its regulatory impact. Based on these positive results, both the US Food and Drug Administration and the European Medicines Agency approved the combination of durvalumab with GC as the new first-line standard of care for aBTCs.

More recently, in 2024, Oh and colleagues published an updated analysis of the pivotal TOPAZ-1 study in The Lancet Gastroenterology & Hepatology (7). The primary endpoint remained OS, with key secondary endpoints including progression-free survival (PFS), objective response rate (ORR), duration of response, disease control rate (DCR), and safety. At the time of the updated data cutoff, the median follow-up was 23.4 months in the experimental group and 22.4 months in the control group. Median OS in the durvalumab plus GC arm was 12.9 months [95% confidence interval (CI): 11.6–14.1], compared to 11.3 months (95% CI: 10.1–12.5) in the placebo plus GC arm, with a hazard ratio (HR) of 0.76 (95% CI: 0.64–0.91). The estimated OS rates at 24 months were 23.6% (95% CI: 18.7–28.9%) versus 11.5% (95% CI: 7.6–16.2%), respectively. Secondary endpoints were also concordantly positive. As previously reported in the interim analysis, median PFS was 7.2 months (95% CI: 6.7–7.4) with durvalumab and 5.7 months (95% CI: 5.6–6.7) with placebo (HR for PFS: 0.75; 95% CI: 0.63–0.89; P=0.001).

Interestingly, the survival benefit associated with durvalumab emerged after 6 months of treatment, as shown by a non-significant piecewise HR for survival in the pre-six-month period (HR =0.91; 95% CI: 0.66–1.25), compared with a significant HR in the post-six-month period (HR =0.71; 95% CI: 0.58–0.88).

A particularly noteworthy aspect of the update was a post-hoc analysis of long-term survivors based on best objective response (progressive disease vs. stable disease vs. partial/complete response). Patients with an OS greater than 18 months were defined as long-term survivors—a population that typically corresponds to the plateau seen in the tail of Kaplan-Meier survival curves. What truly matters in TOPAZ-1 is not the median shift, but the “tail” of the survival curve—those long-term survivors who reshape our expectations for aBTC treatment. Recognizing and characterizing this group should now be a research priority, as these patients hold the key to moving beyond incremental statistics toward meaningful cures. In TOPAZ-1, long-term survivors accounted for 22% (n=153) of the 685 participants: 26% (n=88) in the experimental arm and 19% (n=65) in the control arm. Notably, 44% (n=39) of long-term survivors in the durvalumab group achieved an objective response. Additionally, long-term survivors more frequently presented with relapsed disease at baseline, a neutrophil-to-lymphocyte ratio (NLR) below 3, carbohydrate antigen 19-9 (CA19-9) levels below 500 U/mL, and carcinoembryonic antigen (CEA) levels below 5 ng/mL.

The median duration of treatment was 7.3 months (range, 0.1–31.0 months) in the durvalumab arm. Adverse events (AEs) of any grade occurred in 99% of patients in both treatment arms, with grade 3 or 4 AEs in 74% of patients in the experimental group and 75% in the control group. Treatment discontinuation due to AEs occurred in 13% and 15% of patients, respectively. Immune-related AEs led to treatment discontinuation in only 1% of patients in the durvalumab group, with no treatment-related deaths. The most common grade 3 or 4 AEs across both groups were anemia and neutropenia. Overall, the addition of durvalumab was well tolerated, with no new safety signals.

However, several limitations of the updated TOPAZ-1 study deserve mentioning. This absence of reliable predictive markers is more than a limitation; it is the Achilles’ heel of the study. Without biomarkers, the field risks exposing large patient populations to the toxicity and cost of immunotherapy for a benefit that only a minority will ever experience. While relapse status, NLR, CA 19-9, and CEA levels emerged as potential predictive markers, these were derived from a post-hoc analysis in a limited population and lack sufficient discriminatory power. Identifying robust predictive biomarkers must be a priority to unlock the full potential of immunotherapy in BTC. Moreover, the absence of data on microsatellite instability status or actionable molecular alterations in long-term survivors limits our understanding of potential avenues for subsequent therapeutic strategies. Finally, there is no indicator of the optimal and sufficient duration of maintenance immunotherapy, which was maintained until progression. The delayed separation of survival curves and the high proportion of long-term survivors with stable disease (56%) suggest a potential role for durvalumab in maintenance therapy.

Overall, the positive results of TOPAZ-1 have brought renewed hope for patients with these aggressive and historically refractory cancers. The KEYNOTE-966 phase 3 global trial, consistent with TOPAZ-1, also showed improved median OS [12.7 months (95% CI: 11.5–13.6)] with pembrolizumab (an anti-PD-1 antibody) added to GC, compared to 10.9 months (95% CI: 9.9–11.6) in the placebo plus GC group. However, no statistically significant difference was observed in PFS or ORR (8).

In conclusion, the updated results of TOPAZ-1 highlight that the combination of durvalumab with chemotherapy achieves more than a statistical milestone; it provides the first tangible step forward in a disease long devoid of innovation. Yet, the clinical benefit remains uneven, restricted to a subset of patients whose identity we still fail to define.

Although durvalumab has now been endorsed as a first-line treatment for all patients with aBTC, such blanket recommendations may obscure the reality that only a minority truly benefit.

It would be a misinterpretation to equate regulatory approval with uniform clinical relevance. Instead, the pressing challenge is not whether durvalumab “works” in BTCs, but in whom it works—and whether the modest shift in median survival justifies widespread use without a predictive biomarker strategy. If the field settles for incremental gains, we risk institutionalizing a fragile standard of care. TOPAZ-1 should therefore be viewed as a gateway trial—a proof of concept that demands deeper biological exploration, not only a statistical therapeutic endpoint.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.