Abstract 1748P
Background
Optimal selection of patients (pts) with mUTC who experience long-term benefit with immune checkpoint inhibitors is of high interest. This retrospective analysis evaluated pts with mUTC, who received D in the phase 3b STRONG study following progression on chemotherapy, to explore clinical and laboratory factors associated with clinical outcomes.
Methods
A total of 867 pts (median age 68 yrs, 80% male, 87.1% ECOG PS 0-1) were used for model selection. Univariable analysis of each prognostic factor was done with a subset of pts with non-missing prognostic factor data and non-missing response data to predict objective response rate (ORR) and OS, using PD-L1 expression and previously identified prognostic factors (blood neutrophil/lymphocyte ratio [NLR], visceral metastasis, platelet count, ECOG-PS and LDH). Multivariable analysis included the subset of pts with prognostic factors significantly (P<0.05) associated with response and OS in the univariable analyses. The multivariable model that included all prognostic factors significantly associated with ORR and OS, and that had high predictive ability (via concordance (C)-index), was selected as the final model.
Results
A 4-factor multivariable model with 457 pts (Table), using the combination of PD-L1, NLR, visceral metastases, and platelet count, was associated with improvement in ORR and OS. C-indices of the 4-factor model were 0.69 and 0.66 for ORR and OS, respectively. Table: 1748P
4-factor multivariable model
| Parameter | Group | n | ORR Response, % | OS Events, n | ORR* OR (95% CI) | OS HR (95% CI) |
| PD-L1 expression | Ref: Low | 268 | 13.1% | 181 | 2.57 (1.59, 4.20) | 0.614 ( 0.478, 0.784) |
| Test: High* | 189 | 28.0% | 98 | |||
| NLR | Ref: ≥5 | 120 | 10.8% | 91 | 2.16 (1.17, 4.26) | 0.476 ( 0.370, 0.616) |
| Test: <5 | 337 | 22.3% | 188 | |||
| Baseline visceral metastases | Ref: Yes | 332 | 15.7% | 215 | 1.92 (1.16, 3.16) | 0.655 ( 0.488, 0.866) |
| Test: No | 125 | 28.8% | 64 | |||
| Platelet count | Ref: >1′ULN or >400,000 | 33 | 3.0% | 30 | 4.56 (1.15, 41.34) | 0.387 ( 0.267, 0.581) |
| Test: ≤1′ULN or ≤400,000 | 424 | 20.5% | 249 |
*ORR all factors: P<0.0001PD-L1 assessed using VENTANA PD-L1 (SP263) Assay. High PD-L1 expression defined as ≥25% of tumor cells or ≥25% of tumor-infiltrating immune cells.
Conclusions
In this exploratory retrospective analysis, a 4-factor model composed of readily available factors and comprising tumor PD-L1 expression, visceral metastasis, NLR, and platelet count was associated with improved ORR and OS with D in previously treated pts with progressive mUTC. Further model evaluation in prospective studies is warranted to determine if it may inform selection of pts to optimize therapeutic index.
Clinical trial identification
NCT03084471.
Editorial acknowledgement
Medical writing support, which was in accordance with Good Publication Practice guidelines, was provided by Anne-Marie Manwaring of Parexel (Worthing, UK) and was funded by AstraZeneca.
Legal entity responsible for the study
AstraZeneca.
Funding
AstraZeneca.
Disclosure
G.P. Sonpavde: Financial Interests, Personal, Advisory Board: EMD Serono/Pfizer, Bristol Myers Squibb, Genentech, Merck, Sanofi, Seattle Genetics, AstraZeneca, Exelixis, Janssen, Bicycle Therapeutics, Immunomedics/Gilead, Scholar Rock, G1 Therapeutics, Loxo Oncology, Lucence; Financial Interests, Personal, Advisory Board, Editor of Bladder cancer virtual center of excellence for Practice Update: Elsevier; Financial Interests, Personal, Other, Member of data safety monitoring board: Mereo; Financial Interests, Personal, Other, Author of chapter: Uptodate; Financial Interests, Personal, Invited Speaker: Medscape, Research to practice, Onclive, Physicians Education Resource, Debiopharm; Financial Interests, Institutional, Research Grant: Sanofi, AstraZeneca, Gilead, QED, Predicine, EMD Serono; Non-Financial Interests, Principal Investigator, Steering committee of trial: Bristol Myers Squibb; Non-Financial Interests, Principal Investigator: G1 Therapeutics; Non-Financial Interests, Leadership Role, steering committee of trial: QED; Other, Spouse employment: Myriad Genetics. C.N. Sternberg: Financial Interests, Advisory Role: AstraZeneca, Astellas Pharma, Bayer, Genzyme, Immunomedics, Incyte, Medscape, Merck, MSD, Pfizer, Roche, UroToday. J. Lee: Financial Interests, Personal, Advisory Board: Bristol Myers Squibb, Astellas Korea, AstraZeneca, MSD, Merck; Financial Interests, Personal, Stocks/Shares: Merck, Johnson and Johnson, Amgen, Myovant Sciences, BeiGene, Innovent Biologics, Black Diamond Therapeutics, Zymeworks, Karyopharm Therapeutics; Financial Interests, Institutional, Invited Speaker: Pfizer, Janssen, Novartis, Bristol Myers Squibb, Genetech, Roche, AstraZeneca, MSD, Merck, Bayer Shering Pharma, Seagen, GI Innovation, Amgen. U. Emeribe: Financial Interests, Full or part-time Employment: AstraZeneca. S.M. Oh: Financial Interests, Full or part-time Employment: AstraZeneca. A.R. Dantas De Lima: Financial Interests, Full or part-time Employment: AstraZeneca. S.J. Hotte: Financial Interests, Invited Speaker: Merck, Janssen Oncology, Bayer, Astellas Scientific and Medical Affairs Inc, Eisai; Financial Interests, Advisory Role: AstraZeneca, Bristol Myers Squibb, Ipsen, Roche Canada, Pfizer; Financial Interests, Other: SignalChem, Seattle Genetics, Astellas Pharma, AstraZeneca, Ayala Pharmaceuticals, Bristol Myers Squibb, Clovis Oncology, Janssen Oncology, Merck, Pfizer, Roche/Genentech.