Introduction: The Diagnostic Challenge of Cancer of Unknown Primary
Cancer of Unknown Primary (CUP) remains one of the most challenging diagnoses in clinical oncology, representing approximately 3 to 5 percent of all cancer diagnoses worldwide. By definition, CUP refers to the presence of metastatic disease where the primary site cannot be identified despite a standardized and exhaustive diagnostic workup. Historically, these patients have faced a dismal prognosis, with a median overall survival often ranging between 6 and 9 months. The standard of care has traditionally relied on empiric, platinum-based chemotherapy regimens, which offer limited efficacy due to the biological heterogeneity of the underlying malignancies.
In recent years, the paradigm of CUP management has shifted from a reactive approach to a proactive, molecularly-driven strategy. Advances in next-generation sequencing (NGS), immunohistochemistry (IHC), and gene expression profiling have enabled clinicians to hypothesize the tissue of origin (TOO) or identify actionable molecular alterations. However, translating these technological advances into improved clinical outcomes in a real-world setting requires centralized expertise. This article examines the findings of a retrospective analysis regarding the French national multidisciplinary tumour board (CUP MTB), which offers a blueprint for integrated, evidence-based management of this complex patient population.
Study Design and the French CUP MTB Framework
Launched in 2020, the French national CUP MTB was designed to provide a centralized platform for the coordination of pathological and molecular diagnostic analyses. The board aims to standardize the diagnostic odyssey for CUP patients and provide therapeutic recommendations based on the latest clinical trial data and molecular findings.
This retrospective study analyzed 246 patients referred to the CUP MTB between July 2020 and December 2023. Of these, 187 patients (76 percent) underwent the comprehensive pathological and molecular characterizations recommended by the board. The primary objectives were to evaluate the diagnostic impact—defined as the identification of a putative tissue of origin—and the therapeutic impact, measured by the initiation of MTB-oriented treatments and subsequent overall survival (OS).
Key Findings: Bridging the Diagnostic Gap
Identification of Tissue of Origin
One of the most striking results of the study was the efficacy of the diagnostic workup. In 130 out of 187 patients (70 percent), the integration of clinical data with advanced tumour profiling allowed for the identification of a putative tissue of origin. The distribution of identified primary sites reflected the common culprits in CUP cases: gastrointestinal (22 percent), lung (17 percent), breast (16 percent), and kidney (15 percent). This high rate of identification underscores the value of sophisticated molecular tools when interpreted by an expert centralized body.
Therapeutic Orientation
Beyond diagnosis, the MTB had a profound impact on treatment selection. A total of 149 patients (61 percent) received treatments based on the board’s specific recommendations. Among these, 111 patients (74.5 percent) received what was classified as MTB-oriented treatment. This included systemic therapies directed at the identified tissue of origin (63.8 percent) or targeted therapies directed at specific molecular alterations (10.7 percent). Conversely, 38 patients (25.5 percent) for whom no specific origin or target was identified were treated with standard empiric chemotherapy according to international guidelines.
Survival Outcomes: The Benefit of Precision Medicine
The most critical finding of the analysis was the survival advantage associated with oriented treatment. The median overall survival for patients receiving MTB-oriented therapy was 18.6 months (Interquartile Range [IQR] = 12.0). In comparison, patients who received empiric treatment had a median overall survival of 11.0 months (IQR = 10.5).
Statistical analysis revealed a hazard ratio (HR) of 0.61 (95 percent Confidence Interval [CI] 0.38-0.98, p = 0.04). This represents a nearly 40 percent reduction in the risk of death for those treated with a molecularly or pathologically guided approach. These data suggest that the identification of a tissue of origin is not merely an academic exercise but a clinical necessity that translates into a tangible survival benefit.
Expert Commentary: Interpreting the Real-World Evidence
The results of the French CUP MTB analysis are significant when viewed alongside major clinical trials such as CUPISCO. While randomized controlled trials often operate under idealized conditions, this study demonstrates that the benefits of molecularly guided therapy are achievable in a national, real-life setting.
One of the primary strengths of this study is its demonstration of feasibility. Coordinating a national board for a rare and complex condition requires significant infrastructure, but the 70 percent identification rate of TOO suggests that the investment is justified. However, clinicians must remain cognizant of the limitations. As a retrospective analysis, there is a potential for selection bias; patients referred to a national MTB may have better performance statuses or more accessible tissue for biopsy compared to the general CUP population.
Furthermore, the success of the MTB-oriented approach hinges on the quality of the biopsy material. In many CUP cases, the available tissue is insufficient for extensive NGS or methylation profiling. This study highlights the need for early and adequate tissue acquisition—moving beyond fine-needle aspirations to core-needle biopsies—to ensure that patients can benefit from modern diagnostic pipelines.
From a biological perspective, the survival benefit observed likely stems from the fact that many CUPs are not biologically unique entities but rather common cancers presenting with an atypical, early metastatic phenotype. By identifying these cancers as gastrointestinal, lung, or breast in origin, clinicians can utilize highly effective, site-specific therapeutic sequences that far outperform the generic ‘one-size-fits-all’ empiric chemotherapy.
Conclusion and Future Directions
The retrospective analysis of the French national CUP MTB provides compelling evidence that a centralized, expert-led approach to Cancer of Unknown Primary significantly improves patient outcomes. By integrating clinical, pathological, and molecular data, the board successfully identified the tissue of origin for a majority of patients and directed them toward more effective, personalized treatments.
For the clinical community, the takeaway is clear: the management of CUP should no longer be defined by diagnostic uncertainty and empiric treatment. Instead, referral to dedicated multidisciplinary boards and access to comprehensive molecular profiling should become the standard of care. Future research should focus on further refining molecular signatures and exploring the role of immunotherapy in patients who do not have a clear tissue of origin or targetable mutation, ensuring that no patient is left with limited options.
Funding and clinicaltrials.gov
This study was funded by the Institut Curie and the 2025 French Genomic Medicine Initiative. No clinical trial registration number was provided as this was a retrospective analysis of clinical practice data.