Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session

111P - Application of radiomics signatures and unidimensional vs volumetric measurement of early tumor growth dynamics (TGD) to predict first-line treatment outcomes in patients with stage IV non-small cell lung cancer (NSCLC)

Date

31 Mar 2023

Session

Poster Display session

Presenters

Lawrence Schwartz

Citation

Journal of Thoracic Oncology (2023) 18 (4S): S101-S105.
<article-id>elcc_Ch03

Authors

L. Schwartz1, K. Aggarwal2, D.J. Grootendorst3, S. Kotapati4, M. Fronheiser2, B. Zhao5, C. Coronado-Erdmann2, M. Micsinai-Balan2, M. Karasarides2, A.T. Fojo6, K. Brown2

Author affiliations

  • 1 New York/US
  • 2 Bristol Myers Squibb, Princeton/US
  • 3 Bristol Myers Squibb, 8540 - Princeton/US
  • 4 Bristol Myers Squibb, 08536 - Princeton/US
  • 5 New York Presbyterian Hospital-Columbia University, New York/US
  • 6 Columbia University, New York/US

Resources

Login to get immediate access to this content.

If you do not have an ESMO account, please create one for free.

Abstract 111P

Background

TGD modeling using sum of longest diameters (SLD) is associated with long-term outcomes in NSCLC. Early changes in radiomics features within the tumor may also correlate with survival outcomes. We retrospectively evaluated 3 methods to assess early treatment outcomes: tumor growth rate (g) by SLD, volumetric measurements, and change in radiomics signatures to predict survival outcomes in NSCLC.

Methods

Patients with stage IV NSCLC in CheckMate 9LA treated with first-line nivolumab+ipilimumab+chemotherapy (NIVO+IPI+CHEMO) or CHEMO alone were included. TGD was modeled using radiologically-assessed SLD from ≤5 target lesions or sum of volumes (SVOL) from all measurable lesions >10 mm at baseline, 6, 12, and/or 18 weeks. Measurements were fitted to the TGD model.1 Overall survival (OS) for each growth quartile was estimated by Kaplan–Meier curves. Changes in radiomic features from all measurable lesions >10 mm were assessed at week 6 and 12.

Results

At week 18, low SVOL- and SLD-derived g values were associated with longer median OS across both treatment arms. SVOL-derived g values were more consistent across timepoints if evaluated at week 12 and 18 than SLD-derived g values. Delta radiomics signatures to predict long-term OS at week 6 (table) and 12 performed better than RECIST 1.1 in the NIVO+IPI+CHEMO arm.

Table: 111P

Median OS in groups defined by unidimensional vs volumetric estimates of tumor growth, and by RECIST 1.1 criteria of response vs delta radiomics signature

Median OS, monthsSLD measurementsaSVOL measurementsa
NIVO+IPI+CHEMOCHEMO aloneNIVO+IPI+CHEMOCHEMO alone
g quartile125.817.9.26.419.1
g quartile412.59.5.11.68.2
Median OS, monthsRECIST response (NIVO IPI+CHEMO)Delta radiomics-derived response (NIVO+IPI+CHEMO)
At 6 weeksAt 6 weeks
Progressive disease7.17.4
Stable disease15.014.5
Partial or complete response32.54
a

Four timepoints, week 18. Patients were grouped according to quartiles of g, with quartile 1 representing the subgroup with slowest g.

Conclusions

SVOL-derived g values correlate with longer OS and are more consistent across timepoints than SLD-derived g values at 18 weeks of treatment. Delta radiomics signatures as early as 6 weeks on-treatment were better than RECIST in identifying patients with NSCLC deriving long-term OS benefit. Both findings can potentially inform early decision making in clinical trials and real-world use.

1. Fojo AT et al. J Clin Oncol. 2022;40(16_suppl):Abst 9063.

Clinical trial identification

NCT03215706.

Editorial acknowledgement

Editorial support was provided by Keri Wellington, PhD, and Isobel Markham of Spark Medica Inc.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Disclosure

L. Schwartz: Financial Interests, Personal, Advisory Role: Roche, Novartis; Financial Interests, Personal, Research Grant: Merck, Boehringer Ingelheim. K. Aggarwal: Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb; Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb. D.J. Grootendorst: Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb; Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb. S. Kotapati: Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb; Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb. M. Fronheiser: Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb; Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb. B. Zhao: Financial Interests, Personal, Royalties: Varian Medical Systems; Financial Interests, Institutional, Sponsor/Funding: Bristol-Myers Squibb; Financial Interests, Institutional, Research Grant: National Cancer Institute. C. Coronado-Erdmann: Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb, Incyte. M. Micsinai-Balan: Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb; Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb; Financial Interests, Personal, Other: Bristol-Myers Squibb. M. Karasarides: Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb; Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb. A.T. Fojo: Financial Interests, Personal, Advisory Role: Akita Biomedical; Financial Interests, Personal, Other, Honoraria: Merck; Financial Interests, Institutional, Research Grant: Merck, Ipsen, Pfizer. K. Brown: Financial Interests, Institutional, Full or part-time Employment: Bristol-Myers Squibb; Financial Interests, Personal, Stocks/Shares: Bristol-Myers Squibb.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.