1974P - Improved risk stratification with a prognostic nomogram incorporating the KIT-Variant Allele Frequency (VAF) and the 557/558 deletions: Results from a multi-institutional study in a cohort of intermediate-risk GIST patients

Background

In the intermediate-risk (I-R) GIST patients, the choice of adjuvant imatinib is challenging. Historically, patient risk stratification is based on tumor size, mitotic index, tumor location, and tumor rupture. Innovative measures from NGS, such as the KIT-Variant Allele Frequency (VAF), may act as surrogate of tumor burden, and thus might negatively correlate with prognosis of localized GIST patients, along with specific tumor genotyping.

Methods

We used real-world data from a multicenter, hospital-based, retrospective/prospective cohort study to develop a prognostic nomogram integrating VAF levels and type/gene location of KIT mutations. The receiver operating characteristic (ROC) analysis was used to classify “low” vs “high” VAF values, further normalized on neoplastic cellularity (nVAF). The optimal cut-off was 50% (AUC 0.84; p-value < 0.02). Cox regression/nomogram models were developed in SPSS and R, respectively.

Results

Overall, data from 240 GIST patients treated between 2015 and 2022 at 6 Oncologic Centers in the EURACAN Network were collected. The cohort of I-R patients with tumors harboring KIT mutations, without adjuvant imatinib, included 66 patients. Diameter of primary tumor >5 cm (p=0.005), mitosis >5/mmq (p<0.001), no-gastric site of origin (p=0.01), KIT Exon 11 deletions or delins involving codons 557 and/or 558 (p<0.001), and nVAF > 50% (p=0.006), were significantly associated with Relapse-Free Survival (RFS). At the median follow-up of 24 months, RFS rate of I-R cohort was 74.4% in the high-nVAF group vs 100% in the low-nVAF group. When RFS between the 2 groups was compared, I-R patients with high-nVAF showed poorer RFS than low-nVAF group. Nomogram analysis including KIT-PVs–nVAF, and the presence/absence of 557/558 del/delins, predicted 2-year RFS.

Conclusions

Prognostication in the intermediate-risk subgroup is still challenging for optimal patient selection for adjuvant imatinib. In this sub-population the use of a nomogram incorporating nVAF and type/gene location of KIT mutations, might represent a more accurate tool in the clinical decision-making process.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.