ctDNA-Based Negative Hyperselection May Guide Treatment Selection in mCRC

"Our results suggest that negative hyperselection using a validated and adequately sensitive plasma ctDNA assay may inform appropriate selection of patients for panitumumab treatment regardless of tumor sidedness [left versus right]," according to the study authors.

Negative hyperselection with circulating tumor DNA (ctDNA) testing may highlight patients with metastatic colorectal cancer (CRC) who are suitable to receive frontline therapy with panitumumab (Vectibix) in combination with chemotherapy, according to findings from an exploratory biomarker analysis of the phase 3 PARADIGM trial (NCT02394834) published in Nature Medicine.

Across the overall biomarker evaluable population, the median overall survival (OS) was 35.6 months (95% CI, 31.1-38.9) with panitumumab plus modified 5-fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) compared with 31.6 months (95% CI, 29.3-34.5) in patients who received bevacizumab (Avastin) plus mFOLFOX6 (HR, 0.87; 95% CI, 0.73-1.02). The median OS among patients with no detected gene alterations who thereby met negative hyperselection criteria was 42.1 months (95% CI, 36.4-49.3) vs 35.5 months (95% CI, 31.6-41.4) in each respective arm (HR, 0.76; 0.61-0.95; P = .171).

For those with any gene alteration, the median OS in each respective arm was 24.2 months (95% CI, 18.2-31.0) vs 26.4 months (95% CI, 20.8-30.9) in patients with left-sided primary tumors (HR, 1.08; 95% CI, 0.71-1.64), 14.1 months (95% CI, 11.3-18.7) vs 18.5 months (11.6-25.5) in those with right-sided tumors (HR, 1.33; 95% CI, 0.84-2.11), and 19.2 months (95% CI, 14.8-23.1) vs 22.2 months (95% CI, 19.1-27.7) across the overall population (HR, 1.13; 95% CI, 0.83-1.53).

The median progression-free survival (PFS) in the panitumumab and bevacizumab arms, respectively, was 13.6 months vs 12.8 months in the overall patient population of negative hyperselection (HR, 0.92; 95% CI, 0.75-1.12; P <.001). Moreover, the median PFS in each respective arm was 7.8 months vs 9.8 months among all patients with any gene alteration (HR, 1.68; 95% CI, 1.23-2.29).

Among all negative hyperselected patients, responses occurred in 81.5% (95% CI, 76.2%-86.0%) of those who received panitumumab and 66.8% (95% CI, 60.8%-72.4%) of those who were treated with bevacizumab (OR, 2.19; 95% CI, 1.47-3.29; P <.001). The respective response rates in each arm were 57.8% (95% CI, 48.0%-67.2%) vs 69.1% (95% CI, 58.8%-78.3%) among all patients with gene alterations (OR, 0.61; 95% CI, 0.34-1.09).

“To our knowledge, this is the first report of negative hyperselection using ctDNA in a large phase 3 trial population,” lead study author, Kohei Shitara, MD, of the Department of Gastroenterology and Gastrointestinal Oncology, National Cancer Center Hospital East, Kashiwa, Japan, and coinvestigators wrote. “Our results suggest that negative hyperselection using a validated and adequately sensitive plasma ctDNA assay may inform appropriate selection of patients for panitumumab treatment regardless of tumor sidedness [left versus right].”

In the open-label phase 3 PARADIGM trial (NCT02394795), patients with RAS wild-type unresectable adenocarcinoma originating in the colorectum who received no prior treatment for their disease were randomly assigned 1:1 to receive panitumumab or bevacizumab plus mFOLFOX6. Patients were stratified by study site, age, and whether they had liver metastases.

The trial’s primary end point was OS. Secondary end points included PFS, response, duration of response (DOR), and curative resection rate.

The biomarker analysis of the PARADIGM study included 733 patients overall, with 368 in the panitumumab arm and 365 in the bevacizumab arm. Of 530 negative hyperselected patients, 259 and 271 received panitumumab and bevacizumab, respectively. Additionally, of 203 patients with gene alterations, 109 were in the panitumumab arm, and 94 were in the bevacizumab arm.

Among those with 1 or more gene alterations, the most common included BRAF V600E mutations (10.6%), KRAS mutations (6.0%), and PTEN mutations (5.5%). Additionally, only 1 mutation was reported in 93.0% (n = 106/114) of those with left-sided tumors and 73.8% (n = 62/84) of those with right-sided tumors; 1 mutation was observed in 84.2% (n = 171/203) of all patients with alterations.

The curative resection rate in the overall negative hyperselected population was 18.9% (95% CI, 14.3%-24.2%) with panitumumab vs 11.1% (95% CI, 7.6%-15.4%) with bevacizumab (OR, 1.87; 95% CI, 1.15-3.09). Among all patients with gene alterations, the curative resection rate was 11.0% (95% CI, 5.8%-18.4%) vs 8.5% (95% CI, 3.7%-16.1%) in each respective treatment arm (OR, 1.33; 95% CI, 0.53-3.54).

The median OS was 14.6 months with panitumumab vs 19.8 months with bevacizumab among all patients with a RAS/BRAF mutation or high microsatellite instability (MSI-H; HR, 1.27; 95% CI, 0.88-1.84). Investigators highlighted that median PFS was similar with panitumumab and bevacizumab for patients with RAS/BRAF wild-type disease and microsatellite stable (MSS) disease, although PFS was shorter with panitumumab in those with a RAS/BRAF mutation and MSI-H disease.

Overall, 98.6% of patients in the biomarker population had an adverse effect (AE). Results highlighted comparable rates of any-grade and grade 3 or higher AEs in negative hyperselected patients and those with gene alterations.

Reference

Shitara K, Muro K, Watanabe J, et al. Baseline ctDNA gene alterations as a biomarker of survival after panitumumab and chemotherapy in metastatic colorectal cancer. Nat Med. Published online February 12, 2024. doi:10.1038/s41591-023-02791-w