Abstract 124P
Background
Recent studies suggested that serum parameters measured prior to immune checkpoint inhibitor (ICI) treatment are associated with poor outcomes. It was the aim of this analysis to investigate the correlation of a variety of serum parameters at baseline with the outcome of ICI therapy in patients with solid tumors.
Methods
Data from 114 patients treated with ICPIs between January 2015 to July 2019 at the Medical University of Vienna were collected. Data collection included baseline characteristics, cancer type, serum parameters such as LDH, CRP, albumin and lymphocyte counts (LC) as well as overall survival.. Additionally the Gustave Roussy Immune Score (GRIm score) and the Glasgow prognostic score (GPS) were calculated. Cox regression models including time-dependent effects and strata for tumor type were used. Prognostic factors were pre-selected using a relaxed LASSO approach.
Results
The majority of patients were male (64.9%) and the most common cancer types were non-small cell lung cancer (30.7%) and renal cell carcinoma (21.9%). Elevated LDH and CRP levels at baseline were associated with a shorter median OS (3.72m vs. 13.56m; 7.93m vs. 23.29m; p<0.001 for both). Decreased albumin and LC had a negative impact on OS (15.59m vs. 3.32m; p<0.001 and 16.38m vs. 11.18m, p=0.001). A hypothetical LDH and CRP increase by 20% was associated with a worse OS at 6 months (LDH HR 1.157, 95% CI 1.056 – 1.256; p<0.001; CRP HR 1.059, 95% CI 1.027 – 1.093; p<0.001). Conversely, an assumed increase of albumin or LC led to an improved OS at 6 months (HR= 0.481, 95% CI 0.341 – 0.680; p<0.001 and HR= 0.860, 95% CI 0.787 – 0.941). GRIm Score and GPS were found to be prognostic for OS (GRIm: HR 2.84, 95% CI 1.72 – 4.69; p<0.001 for high vs. low; GPS HR 3.57, 95% CI 1.76 – 7.25; p<0.001 for bad vs. good. In order to compare the performance of our model i.e 20% increase of serum parameters with the above mentioned scores, the proportion of explained variation (PEV) for each score was calculated. GPS and GRIm Score explain 14.6% and 14.8% of the variability in OS, whereas our model has a PEV of 29.5%.
Conclusions
In summary, the results of this analysis suggest that serum parameters might predict the outcome of patients treated with ICI, independently of the tumor type.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
C. Minichsdorfer: Honoraria (self), Travel/Accommodation/Expenses: Merck; Travel/Accommodation/Expenses: MSD; Travel/Accommodation/Expenses: Roche; Travel/Accommodation/Expenses: Janssen; Honoraria (self), Travel/Accommodation/Expenses: BI. M. Schmidinger: Honoraria (self), Travel/Accommodation/Expenses: Roche; Honoraria (self): BMS; Honoraria (self), Travel/Accommodation/Expenses: Pfizer; Honoraria (self): MSD; Honoraria (self): Merck; Honoraria (self): Exelixis; Honoraria (self): Ipsen; Honoraria (self): Alkermes; Honoraria (self), Travel/Accommodation/Expenses: EUSA; Honoraria (self): Eisai. T. Fuereder: Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: MSD; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: Merck; Honoraria (self), Advisory/Consultancy, Research grant/Funding (self): Roche; Honoraria (self), Advisory/Consultancy, Research grant/Funding (institution), Travel/Accommodation/Expenses: BMS; Honoraria (self), Advisory/Consultancy: AstraZeneca; Honoraria (self): Accord; Honoraria (self): Sanofi; Honoraria (self), Advisory/Consultancy: BI; Honoraria (self): Novartis. All other authors have declared no conflicts of interest.