402O - A randomized phase III trial of site-specific therapy guided by the 90-gene expression assay versus empiric chemotherapy in patients with cancer of unknown primary

Background

Empiric chemotherapy remains the standard of care in patients with cancer of unknown primary (CUP). Evidence on site-specific therapy based on multi-gene tissue of origin assay is limited. Here we reported a randomized phase III trial to evaluate the efficacy of site-specific therapy directed by an approved 90-gene expression assay compared with the empirical chemotherapy in CUP patients (NCT03278600).

Methods

De novo patients with CUP not amenable to local radical treatment were randomly assigned (1:1) to receive site-specific therapy or empirical chemotherapy (taxane/gemcitabine plus platinum). The tumor origin in the site-specific therapy group were predicted by the 90-gene expression assay and treatments were administered accordingly. The primary endpoint was progression free survival (PFS) in the intention-to-treat population.

Results

From September 2017 to March 2021, 182 patients were randomly assigned to receive site-specific (N=91) or empiric chemotherapy (N=91). Site-specific therapy guided by molecular assay resulted in more treatment options than the control arm, and the standard chemotherapy regimens for CUP were administered in 26.4% vs. 92.3% (p<0.001) of the population, respectively. At a median follow-up of 42.9 months, the median PFS was significantly longer with site-specific therapy than with empiric chemotherapy (9.6 vs. 6.6 months; HR, 0.68; 95% CI, 0.49-0.93; p=0.017), while favorable OS was observed with site-specific therapy compared with empiric chemotherapy (28.2 vs. 19.0 months) in the intention-to-treat population. Adverse events of grade ≥3 were similar between the two groups (p=0.61).

Conclusions

Site-specific treatment guided by the approved 90-gene-expression assay resulted in more therapy options, significantly improved PFS and favorable OS compared with empirical chemotherapy in de novo CUP patients.

Clinical trial identification

NCT03278600.

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

Q. Xu: Financial Interests, Personal, Affiliate: Canhelp Genomics Co., Ltd. All other authors have declared no conflicts of interest.