Cancers of Unknown Primary Origin: Real-World Clinical Outcomes and Genomic Analysis at the European Institute of Oncology

Luca Boscolo Bielo; Carmen Belli; Edoardo Crimini; Matteo Repetto; Liliana Ascione; Gloria Pellizzari; Celeste Santoro; Valeria Fuorivia; Massimo Barberis; Nicola Fusco; Elena Guerini Rocco; Giuseppe Curigliano

doi:10.1093/oncolo/oyae038

Cancers of Unknown Primary Origin: Real-World Clinical Outcomes and Genomic Analysis at the European Institute of Oncology

Oncologist. 2024 Mar 23:oyae038. doi: 10.1093/oncolo/oyae038. Online ahead of print.

Authors

Luca Boscolo Bielo^{1

2}, Carmen Belli¹, Edoardo Crimini^{1

2}, Matteo Repetto³, Liliana Ascione^{1

2}, Gloria Pellizzari^{1

2}, Celeste Santoro^{1

2}, Valeria Fuorivia^{1

2}, Massimo Barberis⁴, Nicola Fusco^{2

4}, Elena Guerini Rocco^{2

4}, Giuseppe Curigliano^{1

2}

Affiliations

¹ Division of New Drugs and Early Drug Development for Innovative Therapies, European Institute of Oncology, IRCCS, Milan, Italy.
² Department of Oncology and Hemato-Oncology, University of Milan, Milan, Italy.
³ Early Drug Development Service, Memorial Sloan Kettering Cancer Center, New York, USA.
⁴ Division of Pathology, IEO, European Institute of Oncology IRCCS, Milan, Italy.

PMID: 38520742
DOI: 10.1093/oncolo/oyae038

Abstract

Background: Cancer of unknown primary origin (CUP) poses a significant challenge due to poor clinical outcomes and limited treatment options. As such, further definition of clinicopathological factors and genomic profile to better adapt treatment strategies is required.

Methods: Medical records were interrogated to retrospectively include CUP with available clinical and genomics data at the European Institute of Oncology. Next-generation sequencing (NGS) included targeted panels. Statistical analyses were conducted with R Software 4.2.2.

Results: A total of 44 patients were included. With a median follow-up of 39.46 months (interquartile range [IQR] 35.98-47.41 months), median PFS (mPFS) to first-line regimen was 3.98 months (95% CI 3.22-5.98), with a clinical benefit rate of 26% (95% CI 14%-49%), and disease control rate (DCR) limited to 48.28%. Most patients (26 of 31, 83.87%) received platinum-doublet chemotherapy, with no statistically significant difference between first-line treatment regimens. Median OS (mOS) was 18.8 months (95% CI 12.3-39.9), with a 12-month OS rate of 66% (95% CI 50%-85%). All patients received comprehensive genomic profiling (CGP). For 11 patients, NGS was unsuccessful due to low sample quantity and/or quality. For the remaining, TP53 (n = 16, 48%) and KRAS (n = 10, 30%) represented the most altered (alt) genes. No microsatellite instability was observed (0 of 28), while 6 of 28 (21.43%) tumors carried high TMB (≥10 mutation per megabase). Eight of 33 tumors (24.2%) displayed at least one actionable alteration with potential clinical benefit according to ESCAT. Only 2 of them received targeted therapy matched to genomic alterations, with a combined mPFS of 2.63 months (95% CI 1.84-not evaluable) as third-line regimens. Six patients received anti-PD1/PD-L1 therapy, showing a meaningful mPFS of 13 months (95% CI 2.04-not evaluable).

Conclusion: CUP exhibits poor prognosis with limited benefits from standard treatment regimens. A significant proportion of CUPs carry actionable alterations, underscoring the importance of genomic profiling to gather additional treatment opportunities. In addition, immunotherapy might represent a valuable treatment option for a subset of CUP. Finally, accurate definition of sequencing methods and platforms is crucial to overcome NGS failures.

Keywords: cancer of unknown primary; genomic profiling; immunotherapy; treatment options.