Comparing modeling strategies combining changes in multiple serum tumor biomarkers for early prediction of immunotherapy non-response in non-small cell lung cancer.

Q3 Biochemistry, Genetics and Molecular Biology

Tumor Biology Pub Date : 2024-01-01 DOI:10.3233/TUB-220022

Frederik A van Delft, Milou M F Schuurbiers, Mirte Muller, Sjaak A Burgers, Huub H van Rossum, Maarten J IJzerman, Michel M van den Heuvel, Hendrik Koffijberg

{"title":"Comparing modeling strategies combining changes in multiple serum tumor biomarkers for early prediction of immunotherapy non-response in non-small cell lung cancer.","authors":"Frederik A van Delft, Milou M F Schuurbiers, Mirte Muller, Sjaak A Burgers, Huub H van Rossum, Maarten J IJzerman, Michel M van den Heuvel, Hendrik Koffijberg","doi":"10.3233/TUB-220022","DOIUrl":null,"url":null,"abstract":"Background: Patients treated with immune checkpoint inhibitors (ICI) are at risk of adverse events (AEs) even though not all patients will benefit. Serum tumor markers (STMs) are known to reflect tumor activity and might therefore be useful to predict response, guide treatment decisions and thereby prevent AEs.Objective: This study aims to compare a range of prediction methods to predict non-response using multiple sequentially measured STMs.Methods: Nine prediction models were compared to predict treatment non-response at 6-months (n = 412) using bi-weekly CYFRA, CEA, CA-125, NSE, and SCC measurements determined in the first 6-weeks of therapy. All methods were applied to six different biomarker combinations including two to five STMs. Model performance was assessed based on sensitivity, while model training aimed at 95% specificity to ensure a low false-positive rate.Results: In the validation cohort, boosting provided the highest sensitivity at a fixed specificity across most STM combinations (12.9% -59.4%). Boosting applied to CYFRA and CEA achieved the highest sensitivity on the validation data while maintaining a specificity >95%.Conclusions: Non-response in NSCLC patients treated with ICIs can be predicted with a specificity >95% by combining multiple sequentially measured STMs in a prediction model. Clinical use is subject to further external validation.","PeriodicalId":23364,"journal":{"name":"Tumor Biology","volume":" ","pages":"S269-S281"},"PeriodicalIF":0.0000,"publicationDate":"2024-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Tumor Biology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3233/TUB-220022","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Biochemistry, Genetics and Molecular Biology","Score":null,"Total":0}

引用次数: 0

Abstract

Background: Patients treated with immune checkpoint inhibitors (ICI) are at risk of adverse events (AEs) even though not all patients will benefit. Serum tumor markers (STMs) are known to reflect tumor activity and might therefore be useful to predict response, guide treatment decisions and thereby prevent AEs.

Objective: This study aims to compare a range of prediction methods to predict non-response using multiple sequentially measured STMs.

Methods: Nine prediction models were compared to predict treatment non-response at 6-months (n = 412) using bi-weekly CYFRA, CEA, CA-125, NSE, and SCC measurements determined in the first 6-weeks of therapy. All methods were applied to six different biomarker combinations including two to five STMs. Model performance was assessed based on sensitivity, while model training aimed at 95% specificity to ensure a low false-positive rate.

Results: In the validation cohort, boosting provided the highest sensitivity at a fixed specificity across most STM combinations (12.9% -59.4%). Boosting applied to CYFRA and CEA achieved the highest sensitivity on the validation data while maintaining a specificity >95%.

Conclusions: Non-response in NSCLC patients treated with ICIs can be predicted with a specificity >95% by combining multiple sequentially measured STMs in a prediction model. Clinical use is subject to further external validation.

查看原文本刊更多论文

比较结合多种血清肿瘤生物标记物变化的建模策略，以早期预测非小细胞肺癌免疫疗法无应答情况。

背景：接受免疫检查点抑制剂（ICI）治疗的患者有发生不良事件（AEs）的风险，尽管并非所有患者都能从中获益。已知血清肿瘤标志物（STMs）可反映肿瘤活性，因此可用于预测反应、指导治疗决策，从而预防不良反应：本研究旨在比较一系列预测方法，利用多个连续测量的 STMs 预测无应答情况：方法：比较九种预测模型，利用治疗前 6 周测定的 CYFRA、CEA、CA-125、NSE 和 SCC 双周测量值预测治疗 6 个月后的无应答情况（n = 412）。所有方法都适用于六种不同的生物标记物组合，包括两到五种 STM。根据灵敏度评估模型性能，而模型训练的目标是95%的特异性，以确保低假阳性率：结果：在验证队列中，在大多数 STM 组合（12.9% -59.4%）中，在特异性固定的情况下，增强法提供了最高的灵敏度。在验证数据中，CYFRA和CEA的增强达到了最高的灵敏度，同时特异性保持在95%以上：结论：在预测模型中结合多个连续测量的 STMs，可以预测接受 ICIs 治疗的 NSCLC 患者的无应答情况，特异性大于 95%。临床应用有待进一步的外部验证。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Tumor Biology 医学-肿瘤学

CiteScore

5.40

自引率

0.00%

发文量

审稿时长

1 months

期刊介绍： Tumor Biology is a peer reviewed, international journal providing an open access forum for experimental and clinical cancer research. Tumor Biology covers all aspects of tumor markers, molecular biomarkers, tumor targeting, and mechanisms of tumor development and progression. Specific topics of interest include, but are not limited to: Pathway analyses, Non-coding RNAs, Circulating tumor cells, Liquid biopsies, Exosomes, Epigenetics, Cancer stem cells, Tumor immunology and immunotherapy, Tumor microenvironment, Targeted therapies, Therapy resistance Cancer genetics, Cancer risk screening. Studies in other areas of basic, clinical and translational cancer research are also considered in order to promote connections and discoveries across different disciplines. The journal publishes original articles, reviews, commentaries and guidelines on tumor marker use. All submissions are subject to rigorous peer review and are selected on the basis of whether the research is sound and deserves publication. Tumor Biology is the Official Journal of the International Society of Oncology and BioMarkers (ISOBM).