We determined whether sex-associated tissue and blood immune background characterizes surgically resected NSCLC and advanced patients undergoing immunotherapy (IO), potentially affecting clinical outcome. Peripheral blood, collected at surgery from stage I-III, and at baseline and first disease assessment (T1) from (chemo)-IO treated NSCLC, was subjected to FACS analysis of multiple cellular immunophenotypic and functional properties and to multiplex ELISA assay to measure serum cytokines. Tumor Immune Microenvironment (TIME) was assessed by IHC on surgical samples. These parameters were statistically correlated with clinical characteristics. Surgical cohort (n=123; 65 male [M], 58 female [F]): blood immune profile of males was characterized by higher effector (CD8+GnzB+, CD8+PD1+ and CD3-CD56+CD16+ NK) cells and CD14+ monocytes (P < 0.05), while CD4+ and B lymphocytes (P < 0.05) prevailed in females, also exhibiting higher TGFβ1 serum level. Intriguingly, a parallel background was observed at tumor site where increased PD1+ cells (P = 0.04) featured male TIME, whereas CD4 density was higher in women (P = 0.03), also disclosing higher % of PD-L1low cases. Survival outcome appeared to favor female patients. Metastatic cohort (n=153; 103 M, 50 F): significantly higher CD4 number and proliferation and B cells persisted in baseline F blood from advanced cases (P < 0.05). Similarly, a clear trend towards increased cycling or PD1+CD8 and NKs (P = 0.05) was confirmed in M. The CD4- and B-driven immune response in F and the dominant cytotoxic CD8 (P < 0.04) and NK (P = 0.07) effector phenotypes in M were maintained at T1. Moreover, while CD4+CD25+FOXP3high Tregs rose in males following IO, this immunosuppressive counterbalance was eluded in females. A sex-dependent modulation of serum TGFβ1, sPD-L1, IL-6 and TNFα ultimately enclosed a divergent baseline and dynamic cellular/humoral profile underlying the increased incidence (67%) of irAEs in females (P = 0.03). Finally, a trend towards longer median OS and PFS was apparent in IO-treated female NSCLC. The immunity of early and advanced NSCLC patients is trained by gender and might contribute to the heterogeneity of tumor-host interaction and its clinical impact.

The cancer-immune interplay and its role on the evolution of NSCLC might be disclosed through the detection of local and systemic cues. Aim: To determine whether peripheral blood (PB) immune profiles may mirror clinically relevant tumor immune microenvironmental (TIME) features in resected NSCLC patients.

The groundbreaking results of Immune Checkpoint Inhibitors (ICIs) in NSCLC still involve a limited subset of cases, thus imposing an optimization of patient selection. With the aim to non-invasively intercept tumor-host events implicated in cancer immune surveillance and response to immunotherapy, we explored the dynamic of blood immune-inflammatory markers in a cohort of advanced NSCLC treated with first-line ICIs. Peripheral blood was prospectively collected at baseline (T0) and at first radiological disease assessment (T1) from 47 consecutive NSCLC patients undergoing first-line ICI-based therapy. We performed a flow-cytometric analysis of circulating CD3+, CD8+, CD4+, NK, NKT and Tregs as their expression of functional molecules (PD-1, Granzyme B [GnzB], Perforin [Perf]) and proliferative index (Ki67). Soluble PD-L1 (sPD-L1) was determined by immunoassay together with Lung Immune Prognostic Index (LIPI: LDH + derived Neutrophil-to-Lymphocyte Ratio). All these parameters were correlated to objective response rate (ORR) according to RECIST v1.1 criteria. From October 2020 to August 2021, 47 advanced NSCLC patients candidate to receive first-line ICI-based therapy were enrolled. Median age was 67.8 years (range 41-82), 66% were males and 81% underwent chemo-immunotherapy. ORR resulted 51%. Among baseline parameters, number of metastatic sites, bone lesions and poor LIPI negatively correlated with ORR (p<0.05), while a trend towards favorable response was apparent in tissue PD-L1high cases (ORR=67% vs 28% in PD-L1neg). A significant proliferative burst of CD8+PD-1+ lymphocytes carrying cytotoxic molecules (GnzB+, Perf+) coupled with sPD-L1 decline characterized responders. Conversely, CD8+ GnzB+/Perf+ and NK cells dramatically dropped in non-responders (χ2 test, p<0.01). Furthermore, the kinetic and extent of Treg counteraction, likely triggered by the expanding (Ki67+) pool of effector lymphocytes, appeared to be a distinctive feature of responsive patients. Our results suggest that tracking the evolution of blood immune-inflammatory profiles may provide valuable predictors of ICI efficacy in NSCLC patients.

The role of radiotherapy (RT) in immunotherapy-based (IO) combinatory approaches to advanced NSCLC is still uncertain due to its dual immune -suppressive and -stimulatory effect. We performed a longitudinal peripheral blood (PB) analysis to determine whether RT, by affecting immune cell phenotypes and dynamics, impacts on clinical outcome of IO-treated NSCLC patients. PB samples were prospectively collected at baseline (T0) and first disease assessment (T1) on stage IV NSCLC undergoing 1st line IO-based regimens alone (RTnull) or combined with RT (RTpre, within 4 weeks before IO; RTpost, during IO). Flow cytometric analysis included CD3, CD8, CD4, NK, NKT, CD19, CD14 and Treg cells, expression of functional molecules (PD1, Granzyme B [GZB], Perforin [Perf]) and proliferative index (Ki67). PB parameters and their delta variation ([T1-T0/T0] * 100) were correlated with RT administration, Objective Response Rate (ORR) and Progression-free survival (PFS). Among 57 patients, 22 underwent RT either before (RTpre, 32%) or during (RTpost, 68%) IO. RT doses ranged from 8 to 54 Gy according to sites of involvement. No significant differences in IO response and survival emerged between RT and RTnull cases. Compared to RTnull, baseline RTpre immune profiles exhibited increased % of CD8, CD19, CD14 and NK cells expressing PD1, reduced CD4+GZB/Perf+ and Tregs. Delta variation revealed that RTpre attenuated the downregulation of PD1 in CD8, CD4 and CD19 cells following IO, and favored the circulating release of GZB/Perf+ CD8 and CD4. RTpost reduced CD8 number, proliferation and PD1 expression, while increasing NKT. At variance from RTpre, RTpost did not affect PD1+ T cell kinetic, although decreased total and PD1+ NKs. We observed a clear trend towards prolonged PFS in RTpre group (median PFS: RTpre= not reached, RTpost= 7.1 mos, RTnull= 5.9 mos) associated with slightly increased ORR, hinting that the positive cytotoxic (CD8+GZB/Perf+, NK) to suppressive (Tregs) balance triggered by RTpre may result in greater benefit from IO. RT timing may differentially impact on clinical outcome of IO-treated NSCLC patients by shaping immune cells phenotypes and dynamics.

Parallel monitoring of radiomic and blood/tissue immunophenotypic cues may intercept the critical events implicated in Immune Checkpoint Inhibitors (ICIs) efficacy. Thus, we explored radio-immune features and their evolution to provide suitable predictors of ICI response in advanced NSCLC patients. On 58 ICI treated NSCLC cases, we prospectively evaluated: Tumor Immune Microenvironment (TIME) by PD-L1 expression and incidence and spatial distribution of infiltrating lymphocytes (TILs); peripheral blood (PB) at T0 and first disease assessment (T1) for the quantification of effector (NK, CD8, PD1, Granzyme B [GnzB], Perforin [Perf], Ki67) and suppressor (CD4+CD25+FOXP3+ Tregs) phenotypes (flowcytometry), soluble PD-L1 (sPD-L1) and Lung Immune Prognostic Index (LIPI); CT derived Radiomic Features (RFs, n: 851) at T0 and T1. Changes in PB parameters were expressed as Δ% = ([T1 - T0]/T0)*100, while ΔRFs as (T1-T0)/T0. Primary endpoint was tumor response (RECIST v.1.1): CR/PR or SD ≥ 6 months defined clinical benefit (CB), while SD < 6 months or PD non-responders (NR). Baseline immune profile of CB patients comprised: -TIME enriched of CD3+, CD8+ and PD1+ and poor of immune excluded (IE) TILs (Mann-Whitney, p<0.01 vs NR); -PB bearing prominent effector cells, low sPD-L1 and good LIPI, enclosing a highly prognostic model (Fisher, p<0.005). Among PB-TIME associations, PB Tregs correlated directly with IE and inversely with intratumor-CD8+ TILs (p<0.05). Mean Δ% NK, CD8+Ki67+ and CD8+GnzB/Perf+ were, respectively, -20%, -0.4% and -41% in NR and +22%, +170% and +65% in CB, which also displayed a greater boost of counterregulatory Tregs (p<0.05). Out of 657 ΔRFs resulting from pre-processing and Z-score standardization, 11 were differentially regulated in CB vs NR (Mann Whitney, p<0.05). Distinct ΔRFs principal components, encompassing wavelet-HLL_firstorder_Maximum and -HLL_firstorder_Skewness, showed, respectively, direct and inverse correlation with Δ% CD8+Ki67+ lymphocytes. Static and dynamic radio-immune signatures may discern ICI outcome in advanced NSCLC, ultimately enabling tailored therapeutic approaches.

The groundbreaking results of immune checkpoint inhibitors (ICIs) in NSCLC still leave uncovered the identification of prognostic and predictive biomarkers. Radiomics is a non-invasive approach endowed with the potential to unveil clinically relevant clues by decoding tumor characteristics. Thus, we aimed to develop a CT-based radiomic classifier to predict the response to ICIs in advanced NSCLC.

Untangling inter- and intra- tumor heterogeneity functional to clinical decision-making represents an unmet need of the actual Immune checkpoint inhibitors (ICIs)-driven treatment landscape. Multidimensional interrogation of circulating parameters may complement radiomics to non-invasively provide clinically suitable biomarkers. Thus, we aimed at integrating blood hallmarks of systemic inflammation (SI) with high-throughput CT imaging features to develop a predictive model in ICI treated advanced NSCLC.

BackgroundClinically suitable biomarkers to foresee the response to immune checkpoint inhibitors (ICIs) can be achieved by decoding tumor heterogeneity and its evolution during treatment. Thus, we determine whether the longitudinal assessment of radiomic features (RFs) and blood hallmarks of systemic inflammation (SI) may predict ICI efficacy in advanced NSCLC.MethodsOn 92 stage IV NSCLC patients undergoing ICIs, CT-derived RFs and peripheral blood SI parameters, including derived Neutrophil-to-Lymphocyte ratio (dNLR) and lactate dehydrogenase (LDH), were acquired at baseline (T0) and at first disease assessment (T1). Primary endpoint was ICI response per RECIST. CR/PR or SD ≥ 6 months defined clinical benefit (CB) while SD < 6 months or PD non-responders (NR). T1 - T0 delta variations of 852 RFs and 6 SI indices were challenged into machine learning-based predictive models. RFs pre-processing included redundant features elimination and Z-score standardization; L2 penalized logistic regression with Monte-Carlo cross-validation was implemented for wrapper-based feature selection and model training/test. Resulting delta- radiomic (ΔR), immune/inflammatory (ΔI) and integrated (ΔInt) models were compared based on performance metrics.ResultsMedian OS and PFS were 8.1 (95%CI, 4.1-12.2) and 2.6 months (95% CI, 1.1-4.4), respectively; 34 (37%) patients belonged to CB while 58 (63%) were NR. Applying a model validation calibrated at up to 10 parameters, 5 delta-RFs (first- and higher-order classes) and delta-LDH were selected according to ROC-AUC scores and adopted for respective ΔR, ΔI and ΔInt models. Testing the predictive ability of our designed classifiers, ROC-AUC and accuracy (± St.Dev) were 0.86 ± 0.08 and 0.78 ± 0.08 for ΔR, while 0.78 ± 0.09 and 0.67 ± 0.09 for ΔI. The performance of ΔInt model in discriminating ICI response reached ROC-AUC of 0.88 ± 0.07 and accuracy of 0.82 ± 0.08 (P<0.001), thus overtaking that of individual models.ConclusionsWe developed a dynamic blood-radiomic predictor of ICI efficacy in advanced NSCLC suggesting that non-invasive interception of systemic and local events implicated in cancer evolution may implement current predictive models.Legal entity responsible for the studyUniversity Hospital of Parma.FundingHas not received any funding.DisclosureAll authors have declared no conflicts of interest. BackgroundClinically suitable biomarkers to foresee the response to immune checkpoint inhibitors (ICIs) can be achieved by decoding tumor heterogeneity and its evolution during treatment. Thus, we determine whether the longitudinal assessment of radiomic features (RFs) and blood hallmarks of systemic inflammation (SI) may predict ICI efficacy in advanced NSCLC. Clinically suitable biomarkers to foresee the response to immune checkpoint inhibitors (ICIs) can be achieved by decoding tumor heterogeneity and its evolution during treatment. Thus, we determine whether the longitudinal assessment of radiomic features (RFs) and blood hallmarks of systemic inflammation (SI) may predict ICI efficacy in advanced NSCLC. MethodsOn 92 stage IV NSCLC patients undergoing ICIs, CT-derived RFs and peripheral blood SI parameters, including derived Neutrophil-to-Lymphocyte ratio (dNLR) and lactate dehydrogenase (LDH), were acquired at baseline (T0) and at first disease assessment (T1). Primary endpoint was ICI response per RECIST. CR/PR or SD ≥ 6 months defined clinical benefit (CB) while SD < 6 months or PD non-responders (NR). T1 - T0 delta variations of 852 RFs and 6 SI indices were challenged into machine learning-based predictive models. RFs pre-processing included redundant features elimination and Z-score standardization; L2 penalized logistic regression with Monte-Carlo cross-validation was implemented for wrapper-based feature selection and model training/test. Resulting delta- radiomic (ΔR), immune/inflammatory (ΔI) and integrated (ΔInt) models were compared based on performance metrics. On 92 stage IV NSCLC patients undergoing ICIs, CT-derived RFs and peripheral blood SI parameters, including derived Neutrophil-to-Lymphocyte ratio (dNLR) and lactate dehydrogenase (LDH), were acquired at baseline (T0) and at first disease assessment (T1). Primary endpoint was ICI response per RECIST. CR/PR or SD ≥ 6 months defined clinical benefit (CB) while SD < 6 months or PD non-responders (NR). T1 - T0 delta variations of 852 RFs and 6 SI indices were challenged into machine learning-based predictive models. RFs pre-processing included redundant features elimination and Z-score standardization; L2 penalized logistic regression with Monte-Carlo cross-validation was implemented for wrapper-based feature selection and model training/test. Resulting delta- radiomic (ΔR), immune/inflammatory (ΔI) and integrated (ΔInt) models were compared based on performance metrics. ResultsMedian OS and PFS were 8.1 (95%CI, 4.1-12.2) and 2.6 months (95% CI, 1.1-4.4), respectively; 34 (37%) patients belonged to CB while 58 (63%) were NR. Applying a model validation calibrated at up to 10 parameters, 5 delta-RFs (first- and higher-order classes) and delta-LDH were selected according to ROC-AUC scores and adopted for respective ΔR, ΔI and ΔInt models. Testing the predictive ability of our designed classifiers, ROC-AUC and accuracy (± St.Dev) were 0.86 ± 0.08 and 0.78 ± 0.08 for ΔR, while 0.78 ± 0.09 and 0.67 ± 0.09 for ΔI. The performance of ΔInt model in discriminating ICI response reached ROC-AUC of 0.88 ± 0.07 and accuracy of 0.82 ± 0.08 (P<0.001), thus overtaking that of individual models. Median OS and PFS were 8.1 (95%CI, 4.1-12.2) and 2.6 months (95% CI, 1.1-4.4), respectively; 34 (37%) patients belonged to CB while 58 (63%) were NR. Applying a model validation calibrated at up to 10 parameters, 5 delta-RFs (first- and higher-order classes) and delta-LDH were selected according to ROC-AUC scores and adopted for respective ΔR, ΔI and ΔInt models. Testing the predictive ability of our designed classifiers, ROC-AUC and accuracy (± St.Dev) were 0.86 ± 0.08 and 0.78 ± 0.08 for ΔR, while 0.78 ± 0.09 and 0.67 ± 0.09 for ΔI. The performance of ΔInt model in discriminating ICI response reached ROC-AUC of 0.88 ± 0.07 and accuracy of 0.82 ± 0.08 (P<0.001), thus overtaking that of individual models. ConclusionsWe developed a dynamic blood-radiomic predictor of ICI efficacy in advanced NSCLC suggesting that non-invasive interception of systemic and local events implicated in cancer evolution may implement current predictive models. We developed a dynamic blood-radiomic predictor of ICI efficacy in advanced NSCLC suggesting that non-invasive interception of systemic and local events implicated in cancer evolution may implement current predictive models.