The lung-immune prognostic index (LIPI) to predict early mortality among patients affected by advanced solid cancers under immune-checkpoint inhibitors treatments.

Abstract

e14602 Background: The introduction of immune-checkpoint inhibitors (ICI) revolutionized the treatment of several advanced neoplasms. Unfortunately, a considerable rate of patients experiences early death (≤ 90 days). Transversal prognostic factors for early death under ICI single agents or in combinations with other agents are lacking. The lung immune prognostic index (LIPI) includes pretreatment derived neutrophil-to-lymphocyte ratio (dNLR) and lactate dehydrogenase (LDH) levels and demonstrated an affordable prognostic value in advanced solid tumors ICI-treated. Methods: We performed a retrospective study including 498 patients affected by advanced solid malignancies treated with ICI or in combination with other agents. We explored prognostic factors associated with early mortality (≤ 90 days). Then, we tested the LIPI score to predict early mortality risk through Cox-regression models and ROC curve analysis. Results: Most patients included were males (63.2%) with ECOG PS 0-1 (85.9%). Non-small cell lung cancer was the most frequent malignancy (80.3%), followed by head-neck (10.7%), genitourinary (5%), and gastrointestinal (4%) tumors. 324 (68.1%) patients received ICI single agent, while 152 (31.9%) ICI plus other agents (chemotherapy, tyrosine kinase inhibitors), primarily as upfront treatment (60.5%). The LIPI score was assessable for 305 patients. The median overall survival was 8.9 months (95%CI, 7.6-11.6), while the median progression-free survival was 4.5 months (95% 3.8-5.6). In a multivariable model, ECOG PS 2 (HR 1.85, p = 0.013) and intermediate-high LIPI score (HR 2.50, p < 0.001) were associated with increased death risk after adjusting for sex, age, histology, metastatic sites, and line of treatment. In the same model, intermediate-high LIPI score (HR 2.45, p < 0.001) and baseline liver metastasis (HR 1.87, p= 0.006) were associated with an increased risk of disease progression.118 patients (24.8%) experienced early mortality. ECOG PS 2 (HR 3.13, p<0.001 ), >3 metastatic sites (HR 1.93, p 0.018), baseline lung (HR 1.62, p 0.014) and liver (HR 1.74, p=0.01) metastases were associated to increased risk of 90-day death at univariate analysis. An intermediate-high LIPI score was associated with increased death risk in the univariate (HR 6.79, 95% CI, 3.24-14.22, p<0.001) and multivariate (HR 6.84, 95% CI, 2.63-17.80, p<0.001) assessments for 90-day mortality. ICI in combination with other agents was associated with reduced 90-day mortality risk also in the multivariable model (HR 0.46, 95% CI, 0.23-0.93, p=0.029). The Area Under the Curve for the LIPI score was 0.743 (95% CI, 0.68-0.80) for 90-day mortality prediction. Conclusions: We evidenced a considerable risk of early mortality under ICI-based strategies. The LIPI score predicted 90-day mortality risk regardless of the treatment and histotype.