Integrating clinico-pathological features and longitudinal liquid biopsy results to predict clinical benefit from immune checkpoint Inhibitors in advanced Non-Small Cell Lung Cancer patients

Abstract

Background: Immune checkpoint inhibitors (ICIs) are widely used in non-oncogene addicted Non-small cell lung cancer and characterized by high heterogeneity in clinical benefit. Different combination strategies are available in first-line setting and PD-L1 is the only predictive marker used in clinical practice.
Methods: Patients with NSCLC treated with ICI single-agent according to clinical practice were prospectively enrolled. Liquid biopsy was performed at the time of first dose administration (T1), after 3 weeks (T2) and at the time of radiological evaluation (T3). Cell free DNA (cfDNA) was quantified (ng/ml) by qPCR and analysed by NGS targeted panel.
Molecular variables used for association with outcome endpoints were: cfDNA quantification as static parameter, dynamic cfDNA change (DT2-T1), variant allele frequency of the gene with the highest frequency at base line (MaxVAF) and dynamic maxVAF change (DT2-T1). Cox regression analysis was used to build integrated predictive models.
Results: 113 patients were included. At multivariate analysis, PD-L1 negativity, T1 cfDNA, cfDNA increase (DT2-T1), and maxVAF at T2 were significantly associated with shorter PFS; PD-L1 negativity, squamous histology, T1 cfDNA, increase of cfDNA (DT2-T1), and maxVAF at T2 were significantly associated with worse OS. Integrated model permitted to build a nomogram and establish three groups of patients deriving different clinical benefit from ICI. The model was tested in patients expressing PD-L1 ³50% and treated with first-line pembrolizumab (n=57) and was able to identify elevated maxVAF at T2 and increase (DT2-T1) of cfDNA as independently associated with worse PFS; higher levels of maxVAF at T2 and increase (DT2-T1) of cfDNA with worse OS. Derived integrated model was able to identify patients with different clinical benefit (high, intermediate, low risk).
Conclusions: We developed an integrated nomogram to stratify NSCLC patients deriving different clinical benefit from ICIs which outperforms individual predictive markers.

Full text

Page 1/20
Integrating clinico-pathological features and
longitudinal liquid biopsy results to predict clinical
benet from immune checkpoint Inhibitors in
advanced Non-Small Cell Lung Cancer patients
Andrea Boscolo Bragadin
Veneto Institute of Oncology IOV – IRCCS https://orcid.org/0000-0001-9614-0814
Paola Del Bianco
Veneto Institute of Oncology IOV - IRCCS https://orcid.org/0000-0002-6597-0265
Elisabetta Zulato
Veneto Institute of Oncology IOV – IRCCS
Ilaria Attili
Veneto Institute of Oncology IOV – IRCCS
Alberto Pavan
Veneto Institute of Oncology IOV - IRCCS https://orcid.org/0000-0003-4695-0100
Jessica Carlet
Veneto Institute of Oncology IOV – IRCCS
Ludovica Marra
Veneto Institute of Oncology IOV – IRCCS
Valentina Guarneri
Veneto Institute of Oncology IOV – IRCCS
Stefano Indraccolo
Veneto Institute of Oncology IOV – IRCCS
Laura Bonanno (  laura.bonanno@iov.veneto.it )
Veneto Institute of Oncology IOV - IRCCS https://orcid.org/0000-0001-5218-4970
Article
Keywords: Primary resistance, immunotherapy, NGS, ctDNA, predictive biomarkers, liquid biopsy
Posted Date: January 25th, 2024
DOI: https://doi.org/10.21203/rs.3.rs-3848506/v1
License:   This work is licensed under a Creative Commons Attribution 4.0 International License. 
Read Full License

Page 2/20
Additional Declarations: (Not answered)

Page 3/20
Abstract
Background: Immune checkpoint inhibitors (ICIs) are widely used in non-oncogene addicted Non-small
cell lung cancer and characterized by high heterogeneity in clinical benet. Different combination
strategies are available in rst-line setting and PD-L1 is the only predictive marker used in clinical
practice.
Methods: Patients with NSCLC treated with ICI single-agent according to clinical practice were
prospectively enrolled. Liquid biopsy was performed at the time of rst dose administration (T1), after 3
weeks (T2) and at the time of radiological evaluation (T3). Cell free DNA (cfDNA) was quantied (ng/ml)
by qPCR and analysed by NGS targeted panel.
Molecular variables used for association with outcome endpoints were: cfDNA quantication as static
parameter, dynamic cfDNA change (DT2-T1), variant allele frequency of the gene with the highest
frequency at base line (MaxVAF) and dynamic maxVAF change (DT2-T1). Cox regression analysis was
used to build integrated predictive models.
Results: 113 patients were included. At multivariate analysis, PD-L1 negativity, T1 cfDNA, cfDNA increase
(DT2-T1), and maxVAF at T2 were signicantly associated with shorter PFS; PD-L1 negativity, squamous
histology, T1 cfDNA, increase of cfDNA (DT2-T1), and maxVAF at T2 were signicantly associated with
worse OS. Integrated model permitted to build a nomogram and establish three groups of patients
deriving different clinical benet from ICI. The model was tested in patients expressing PD-L1 ³50% and
treated with rst-line pembrolizumab (n=57) and was able to identify elevated maxVAF at T2 and
increase (DT2-T1) of cfDNA as independently associated with worse PFS; higher levels of maxVAF at T2
and increase (DT2-T1) of cfDNA with worse OS. Derived integrated model was able to identify patients
with different clinical benet (high, intermediate, low risk).
Conclusions: We developed an integrated nomogram to stratify NSCLC patients deriving different clinical
benet from ICIs which outperforms individual predictive markers.
BACKGROUND
The introduction of immunotherapy in the treatment of advanced Non-small cell lung cancer (NSCLC) is
one of the most important advances for patients’ outcome and quality of life 1,2.
Currently, most of the patients diagnosed with advanced NSCLC receive treatment with immune
checkpoint inhibitors (ICIs) with important exception represented by so-called oncogene addicted
diseases 1.
Different treatment strategies including ICIs are currently available for rst-line treatment: anti PD-1/PD-
L1 monotherapy, combination of chemotherapy and anti PD-1 antibody, anti PD-1 plus anti-CTLA4 plus
short course of chemotherapy 1,3–9. Nowadays the medical choice among treatment options is based on

Page 4/20
PD-L1 expression, the only available predictive biomarkers in clinical practice, and toxicity proling 1.
Great heterogeneity in response and duration of response has been observed. In particular, a subgroup of
patients could even derive detrimental effects from immunotherapy single agent, while 15–20% of
patients can have prolonged clinical benet and are dened as long-survivors, even in the absence of
more toxic combination strategies 10–12.
To shed light on heterogeneity in clinical benet, several biomarkers have been investigated and results
support the concept that response to immunotherapy is likely not related to a single biomarker but rather
to the dynamic interaction among factors related to host, tumor and microenvironment 13,14. We thus
hypothesized potential role of circulating tumor DNA (ctDNA) as surrogate of tumor burden and early
treatment-related biological changes in order to anticipate response and clinical benet following
ICIs12,15,16.
We performed a prospective study in advanced NSCLC patients treated with single-agent ICIs and
integrated clinico-pathological features and results of ctDNA monitoring in order to build a predictive
model.
METHODS
Patients’ enrolment and study design
Patients with advanced NSCLC and treated with ICIs at the Istituto Oncologico Veneto – IOV IRCCS
(Padova, Italy) were enrolled as part of the prospective study called MAGIC between January 2017 and
August 2019 12,15. Main inclusion criteria were histological diagnosis of EGFR-ALK-ROS1 wild type
NSCLC, feasibility of ICI treatment according to clinical practice, the possibility of an appropriate clinical
and radiological follow-up and signed informed consent.
According to Italian regulatory indications, ICIs were administered after progression to chemotherapy
independently of PD-L1 expression between January and July 2017. Since July 2017 patients were
treated with rst-line pembrolizumab in the presence of PD-L1 expression in at least 50% of tumor cells in
tumor biopsies and with second-line ICIs (nivolumab, pembrolizumab, atezolizumab) in the presence of
PD-L1 inferior to 50%.
Liquid biopsies were collected at pre-planned time points during ICIs: at the same day of rst
administration of treatment (baseline or T1), after 2 or 3 weeks according to treatment schedule (T2), at
the time of rst radiological evaluation (T3).
Radiological and clinical follow-up were performed according to clinical practice.
The ethics committee of Istituto Oncologico Veneto evaluated and approved study design and informed
consent (2016/82, 12/12/2016). Written informed consent was obtained from all patients before study
entry. The study was conducted in accordance with the precepts of the Helsinki declaration.

Page 5/20
Plasma sample collection
Blood draws (~ 20 ml) were collected at each time point in two cell-free DNA BCT tubes (Streck Corporate,
La Vista, NE, USA) and processed for plasma extraction within the next 24-72h. As previously described
12,15, two sequential centrifugations were performed to obtain plasma suitable for NGS analyses: rst,
blood draws were centrifuged at 2000 × g for 10 min at 4°C to separate plasma, then plasma was
centrifuged at 20,000 × g for 10 min at 4°C to pellet cellular debris and supernatant was stored at − 80°C,
until cfDNA extraction.
cfDNA extraction and quantity/quality assessment
cfDNA was extracted from 3–5 ml of plasma, using QIAamp Circulating Nucleic Acid Kit (Qiagen, Venlo,
Netherlands) following manufacturer’s instructions. Circulating nucleic acids were eluted in 30 µl of
nuclease free water. Quantity and amplicability of cfDNA was assessed by qPCR included in the
Myriapod NGS-IL 56G Onco panel CE IVD kit (Diatech Pharmacogenetics, Jesi, Italy).
cfDNA library preparation and sequencing
Targeted Next Generation Sequencing (NGS) was performed in liquid biopsies collected at each time-
point using Myriapod NGS-IL 56G Onco panel CE IVD kit (Diatech) covering hot spot regions of 56
frequently mutated tumour suppressor genes and oncogenes in cancers (Supplementary Table1). NGS
libraries were prepared starting from 1.27 ng of cfDNA to a maximum of 25 ng. Eighteen Indexed libraries
were pooled together, diluted at a nal concentration of 12 pM, and sequenced on an Illumina MiSeqDx
System using the 600 cycles MiSeq Reagent Kit v3 (Illumina) in pair-end mode (2 x 151 cycles). FastQ
les obtained from sequencing were analysed using Myriapod NGS Data Analysis Software.
Synonymous and intronic variants were excluded from analyses. Remaining mutations were consulted
on Varsome to dene their pathogenicity according to ACMG rules 17. Only pathogenic or likely
pathogenic mutations and variants of uncertain signicance (VUS) were considered for longitudinal
analyses.
Statistical Analyses
Radiological response (RR), progression free survival (PFS), and overall survival (OS) were the primary
endpoints considered for the analysis.
Given the exploratory nature of the study, there was no formal hypothesis or power sample size
calculation.
RR was graded as complete response (CR), partial response (PR), stable disease (SD), and progressive
disease (PD) according to RECIST 1.1 criteria18. PFS was calculated as the time from the beginning of
systemic treatment to the occurrence of radiological disease progression or death from any cause. OS
was calculated as the time from the beginning of systemic treatment to death from any cause. Patients

Page 6/20
who did not develop a survival event during the study period were censored at the date of the last
observation.
As molecular variables, we considered both cfDNA concentration in plasma (ng/ml) and VAF. In the
presence of multiple mutations, the one with the highest VAF (maxVAF) at baseline was selected as the
reference for NGS analysis.
Continuous variables were described as median and interquartile range and their distributions according
to different groups were compared using the Kruskal-Wallis test. Categorical variables were described as
counts and percentages and compared between groups using the χ2 or Fisher exact test as appropriate.
The median follow-up time was based on the reverse Kaplan-Meier estimator.
The correlation between cfDNA and maxVAF at each time point was calculated by using the Spearman's
rank correlation coecient.
Each biomarker was analysed for association with clinical outcome as a categorical variable according
to high and low levels. Optimal cut-points were estimated by maximizing the discriminative ability of the
Cox model for PFS and OS, and validated with bootstrapping.
Main effects of candidate clinical and molecular predictors were included in multiple Cox proportional
hazards regression models for the survival outcome. No deviation from the proportional hazards
assumption was found by the test statistic of Grambsch and Therneau.
A backward elimination with the Bayesian Information Criterion (BIC) was applied for selecting all
variables independently associated with the outcome. Results were displayed in terms of hazard ratios
together with 95% CI. A nomogram of the nal reduced model was generated for predicting the 6- and 12-
month survival probabilities.
The ability of the models to differentiate between high-risk patients and low-risk patients, was calculated
with Harrell's concordance (C) index, adjusted through 1000 bootstrap resamples.
Lastly, patients were stratied into three risk groups according to optimal cut-points of their individual
linear predictor score, and the log-rank test was used to compare groups.
The nonparametric Kaplan-Meier method was used to estimate the survival probabilities and median time
was provided along with the corresponding 95% condence interval (CI) estimated using the Brookmeyer-
Crowley method.
All statistical tests used a two-sided 5% signicance level. Statistical analyses were performed using
RStudio (RStudio: Integrated Development for R. RStudio, Inc., Boston, MA).
RESULTS

Page 7/20
Patients, treatments and outcome
Among 149 advanced NSCLC patients complying clinical inclusion criteria, 113 (76%) were considered as
evaluable for cfDNA studies (Table1 and Supplementary Table2), based on the availability of complete
clinico-pathological data. Main clinical features of the study population are summarized in Table1.
Table 1
Clinical features of the study population. Included patients refers to patients included in
multivariate analyses, with all clinic/pathological data available
  Included patients* (N = 113, 76%) Total (N = 149)
Age (years) Median (Q1-Q3) 67.0 (62.0, 73.0) 67.0 (62.0, 74.0)
Sex Male 74 (65.5%) 97 (65.1%)
 Female 39 (34.5%) 52 (34.9%)
Smoking No 13 (11.5%) 19 (12.8%)
 Former 61 (54.0%) 79 (53.0%)
 Yes 39 (34.5%) 51 (34.2%)
ECOG-PS 0 48 (42.5%) 69 (46.3%)
 1–2 65 (57.5%) 80 (53.7%)
Histology Non-squamous 98 (86.7%) 129 (86.6%)
 Squamous 15 (13.3%) 20 (13.4%)
PD-L1 Negative 38 (33.6%) 38 (25.5%)
 Positive 75 (66.4%) 75 (50.3%)
 Not available - 36 (24.2%)
Extrathoracic sites No 44 (38.9%) 62 (41.6%)
 Yes 69 (61.1%) 87 (58.4%)
N metastatic sites 0–1 59 (52.2%) 74 (49.7%)
 2–4 54 (47.8%) 75 (50.3%)
Treatment setting 1L 57 (50.4%) 58 (38.9%)
 > 1L 56 (49.6%) 91 (61.1%)
Follow-up (months) Median (Q1-Q3) 42.2 (39.5, 48.7) 44.3 (40.5, 52.9)
Fifty-seven patients were treated with rst-line pembrolizumab, while 56 were treated with ICIs following
progression to rst line chemotherapy. Overall, partial response rate was 27.4% while 43 (38.1%) patients

Page 8/20
had progressive disease (PD) as best response. Disease control (PR + SD) was achieved in 37 patients
out of 57 treated in rst-line (Table2). In the overall study population, with a median follow-up of 42.2
months, median PFS was 6.3 months (95%CI: 5.3, 8.1), and median OS was 12.0 months (95%CI: 8.1,
14.6) (Table2).
Table 2
Outcome of the study population, in terms of radiological response (RR), progression-free survival (PFS)
and overall survival (OS).
  1 treatment line
(N = 57) >1 treatment line
(N = 56) Total (N =
113)
p
value
Radiological
response PR 21 (36.8%) 10 (17.9%) 31 (27.4%)
0.0690
 SD 16 (28.1%) 23 (41.1%) 39 (34.5%) 
 PD 20 (35.1%) 23 (41.1%) 43 (38.1%) 
PFS (months) median
(95% CI) 8.1 (5.2, 11.3) 5.8 (4.3, 7.3) 6.3 (5.3,
8.1)
0.0900
OS (months) median
(95% CI) 13.1 (8.9, 29.5) 10.8 (6.8, 14.5) 12.0 (8.1,
14.6) 0.0200
In the whole study population, 17 patients died within 12 weeks from the start of ICIs and 2 cases comply
radiological criteria for hyper-progression10,12,19.
Among clinico-pathological features evaluated, smoking status and PD-L1 levels were able to affect
probability of PFS with statistical signicance at univariate analysis, as shown in Supplementary Table3.
Factors able to affect OS included levels of PD-L1, and receiving immunotherapy as rst line treatment
(Supplementary table 4).
Results of longitudinal liquid biopsy
Supplementary Table2 summarises results of longitudinal liquid biopsies. Median ctDNA quantication
at baseline was 9.91 ng/ml (Q1-Q3: 6.64–15.21, Table3a). 72 out of 113 patients (64%) had detectable
mutations at baseline. Of 41 patients (36%) with no detectable mutations at baseline, 8 were found
positive at T2 and 2 at T3 (Supplementary Table2). The concentration of cfDNA assessed at each time-
point was signicantly correlated with the levels of maxVAF detected (Supplementary Table5).
We considered basal liquid biopsy results in order to test potential association with clinical features and
tumor burden (Supplementary Table6). Higher cfDNA concentration was detected among former smoker
patients (p = 0.025), whereas no association was observed with tumor burden related factors. On the
other side, patients with two or more metastatic sites have signicantly higher maxVAF (p = 0.005).
Liquid biopsy and outcome

Page 9/20
Baseline (T1) and T2 cfDNA and maxVAF were signicantly associated with RR, in particular patients
experiencing PD as best radiological response had higher cfDNA and higher maxVAF compared to those
experiencing SD or PR (Table3a).
At univariate survival analysis, higher cfDNA at baseline and T2 and maxVAF at baseline and T2, as well
as increase of cfDNA (ΔT2-T1) and of maxVAF (ΔT2-T1) from baseline were signicantly associated with
the risk of progression and shorter survival (Supplementary Table7–8). Biomarkers were then
categorized according to median in-bag cut-point values from bootstrap validation to identify cut-offs
dening high versus low for each variable that was used for subsequent analyses (Supplementary
Tables9–10).
We also considered potential role of liquid biopsy performed at the time of radiological response
assessment (T3) to further stratify outcome of patients achieving radiological response (Table3b-c). In
particular median PFS was 6.6 (95% CI: 5.5–8.6) months for patients with high maxVAF at T3,
versus
12.2 (95% CI: 9.8–17.6) months in the presence of low maxVAF at T3 (p: 0.003, Table3b) and median OS
was 13.3 (95% CI: 7.3–21.2) and 28.5 (95% CI: 19.9–35.7) months in the presence of high or low maxVAF
at T3, respectively (p: 0.016, Table3c).

Page 10/20
Table 3
Liquid biopsy results and radiological response a) Association of cfDNA quantication and maxVAF at
T1 and T2 with Radiological response
  PR (N = 31) SD (N = 39) PD (N = 43) Total (N =
113)
p
value
maxVAF
at T1 N 31 39 43 113 0.0050
 Median
(Q1, Q3) 1.23 (0.02,
5.24) 0.02 (0.02,
2.25) 4.61 (0.42,
15.02) 1.28 (0.02,
7.65) 
maxVAF
at T2 N 29 37 39 105 <
0.001
 Median
(Q1, Q3) 0.02 (0.02,
1.02) 0.02 (0.02,
1.39) 3.81 (1.52,
13.66) 0.99 (0.02,
3.91) 
maxVAF
at T3 N 29 38 16 83 0.0010
 Median
(Q1, Q3) 0.02 (0.02,
0.02) 0.02 (0.02,
1.43) 4.84 (0.02,
10.79) 0.02 (0.02,
1.66) 
cfDNA at
T1 N 31 39 43 113 <
0.001
 Median
(Q1, Q3) 8.81 (5.49,
12.34) 8.28 (6.36,
10.90) 13.80 (10.08,
24.34) 9.91 (6.64,
15.21) 
cfDNA at
T2 N 29 37 39 105 <
0.001
 Median
(Q1, Q3) 8.50 (4.84,
13.01) 8.22 (5.76,
12.86) 14.97 (10.56,
51.69) 10.81 (6.10,
17.88) 
cfDNA at
T3 N 29 38 16 83 <
0.001
 Median
(Q1, Q3) 6.10 (4.99,
8.15) 10.06 (6.59,
12.51) 16.66 (10.71,
32.31) 8.25 (5.30,
13.08) 
b) Impact on progression free survival of molecular variables performed at time of radiological
assessment (T3) for patients achieving RR (Univariate analysis)

Page 11/20
  PFS Univariate analysis
  n events Median PFS
(95%CI)
logrank
HR(95%CI)
p
value
cfDNA at T3 < 9.16 43 31 15.4 (10.0;24.9)
0.0013
Ref 
 >=9.16 24 22 7.8 (5.8;10.4)  2.46
(1.40;4.33) 0.0017
maxVAF at
T3 < 1.13 53 40 12.2 (9.8;17.6)
0.0028
Ref 
 >=1.13 14 13 6.6 (5.5;8.6)  2.56
(1.35;4.84) 0.0038
c) Impact on overall survival of molecular variables performed at time of radiological assessment (T3)
for patients achieving RR (Univariate analysis)
  OS Univariate analysis
  n events Median (95%CI)
logrank
HR(95%CI)
p value
cfDNA at T3 < 5.27 20 10 35.7 (14.5;NA)
0.0099
Ref 
 >=5.27 47 36 19.9 (12.9;25.8)  2.47 (1.22;5.02) 0.0123
maxVAF at T3 < 1.03 52 33 28.5 (19.9;35.7)
0.0158
Ref 
 >=1.03 15 13 13.3 (7.3;21.2)  2.19 (1.14;4.19) 0.0179
Integrated model of liquid biopsy results and clinical
features
At multivariate analysis, including clinico-pathological and molecular features, PD-L1 negativity, baseline
cfDNA, increase of cfDNA (ΔT2-T1), and maxVAF at T2 were signicantly associated with shorter PFS
(Table4); moreover, PD-L1 negativity, squamous histology, baseline cfDNA, increase of cfDNA (ΔT2-T1),
and maxVAF at T2 were signicantly associated with worse OS (Table5).

Page 12/20
Table 4
Multivariate analysis including clinico-pathological and molecular variables and their impact
on progression-free survival
PFS Multiple analysis Shrunked 
  HR(95%CI)
p value
HR(95%CI)
p value
PD-L1 Negative ref  Ref 
 Positive 0.47 (0.30, 0.73) 0.0009 0.51 (0.32, 0.79) 0.0030
cfDNA at T1 < 7.78 Ref  Ref 
≥ 7.78 1.65 (1.06, 2.59) 0.0272 1.57 (1.01, 2.46) 0.0480
cfDNA ∆T2-T1 <-0.11 Ref  Ref 
≥-0.11 2.07 (1.31, 3.27) 0.0017 1.92 (1.22, 3.44) 0.0050
maxVAF at T2 < 1.94 Ref  Ref 
≥ 1.94 2.43 (1.57, 3.77) 0.0001 2.22 (1.43, 3.44) 0.0004
C-index  0.711  0.699 
Table 5
Multivariate analysis including clinico-pathological and molecular variables and their impact on
overall survival
OS Multiple analysis Shrunked 
  HR(95%CI)
p-value
HR(95%CI)
p-value
PD-L1 Negative Ref  Ref 
 Positive 0.43 (0.27, 0.69) 0.0005 0.49 (0.31, 0.78) 0.0026
Histology Non-squamous Ref  Ref 
 Squamous 3.34 (1.74, 6.42) 0.0003 2.83 (1.47, 5.43) 0.0018
cfDNA at T1 < 7.78 Ref  Ref 
≥ 7.78 1.67 (1.05;2.65) 0.0308 1.55 (0.98;2.47)
0.0628
cfDNA ∆T2-T1 < 3.8 Ref  Ref 
≥ 3.8 1.85 (1.15, 2.98) 0.0106 1.70 (1.06, 2.73) 0.0277
maxVAF at T2 < 1.94 Ref  Ref 
≥ 1.94 2.70 (1.70, 4.29) 0.0001 2.35 (1.48, 3.73) 0.0003
C-index  0.737  0.723 

Page 13/20
Supplementary Figs.1 and 2 show related nomogram to assess risk factors and potentially anticipate
outcome following immunotherapy in clinical practice. Based on each patient's total score in the
nomogram, patients were stratied into three risk groups in order to early identify patients with high
probability of rapid progression and dismal outcome, patients with intermediate clinical benet and
patients likely to achieve durable clinical benet (Fig.1A-B; Supplementary Table11–12). In particular,
median PFS and OS for low-risk group were 23.1 (95% CI: 10-ns) and 30.1 (95% CI: 28.4-NS) months,
median PFS and OS for intermediate group were 6.4 (95% CI: 5.6–9.8) and 12.1 (95% CI: 8.9–21.1)
months; patients in the higher risk group had median PFS of 2.5 (95% CI: 1.9–4.3) and median OS of 4.2
(95% CI: 3.3–7.8) months.
Integrated model and rst line treatment
Subsequently, we tried to validate our integrated prediction model in a more homogeneous clinical
setting.
Fifty-seven patients treated with rst-line pembrolizumab were included in the analysis. Median PFS was
8.1 months (95%CI: 4.6–11.3), median OS was 13.1 months (95%CI: 8.0-28.2). Twenty-one (36.8%)
patients achieved a PR, while 20 patients experienced PD as best response (Table2).
In the multiple Cox model, high maxVAF at T2 (HR = 2.5, 95%CI: 1.3─4.6), and an increase (ΔT2-T1) of
cfDNA (HR = 2.3, 95%CI: 1.2─4.2) remained signicantly associated with a worse PFS (Supplementary
Table11). Similarly, high levels of maxVAF at T2 (HR = 2.5, 95%CI: 1.3─4.7) and an increase (ΔT2-T1) of
cfDNA (HR = 2.3, 95%CI: 1.2─4.3) remained signicantly associated with a worse OS (Supplementary
Table13).
Figure 2 depicts three-group risk stratication model based on nomogram created for rst-line treatment.
In particular, median OS was 1.8 (95% CI: 1.2-NA) months among patient considered at high risk, 6.5 (95%
CI: 5.5–21.2) months among patients considered as intermediate risk and 29.5 (95% CI: 14.5-NA) among
patients considered at low risk (Fig.2, Supplementary 14–15).
Among patients achieving RR according to RECIST V1.1, the results of liquid biopsy performed at the time
of radiological evaluation (T3), was able to anticipate long-term clinical benet (Supplementary
Table12–13).In particular, high maxVAF at T3 was associated with a median PFS of 7.0 (95% CI: 4.6-NA)
months and a median OS of 12.8 (95% CI: 6.5-NA) months versus a median PFS of 16.9 (95% CI: 10.3–
37.9) months and a median OS of 35.7 (95% CI: 21.6-NA) for patients with low maxVAF at T3
(Supplementary Table16–17)
Discussion
Predicting clinical benet of immunotherapy in advanced NSCLC is recognized as one of the main
challenges for clinical and translational research in oncology.

Page 14/20
The potentialities of using dynamic liquid biopsy during treatment with ICIs have already been discussed
elsewhere 16,20–23. Some data have already been published showing how changes in ctDNA during
treatment with ICIs might be associated with outcome in advanced NSCLCs and other solid tumors
12,15,24–27. Even though the idea of early monitoring surrogate markers of tumor burden in blood during
treatment to anticipate radiological evaluation and long-term clinical benet is increasingly accepted by
the scientic community, available data are characterized by high heterogeneity in methods, on-treatment
time-point evaluation, patients population and clinical end-points. A recent analysis pooled and
harmonized data from 200 patients included in ve different trials and showed that predictive role of
dynamic analysis of ctDNA can be conrmed28. Nevertheless a practical model to be tested in a
interventional clinical trial and afterwards applied in clinical practice is still awaited.
In our study, we prospectively included a single-centre population of advanced NSCLC patients, treated
according to clinical practice (outside interventional clinical trial). All the patients performed liquid biopsy
at the pre-established time-points and we included into our integrated analysis only patients with all
clinico-pathological data available, including PD-L1 expression in tissue at baseline and lack for driver
alterations, commonly tested in clinical practice in tissue.
As far as molecular variables are concerned, we rst tested multiple candidate (mean and median VAF,
maxVAF, absolute and relative changes in ctDNA and VAF assessed by different methods), which were all
consistent in association with clinical outcome endpoints (data not shown) and selected the one most
likely to be applicable to clinical practice (cfDNA and maxVAF as static parameter and absolute change in
ctDNA and maxVAF as dynamic parameter). Consistently, we decided to use a targeted NGS kit including
a relatively small number of genes in order to focus on quantitative variations in plasma and increase
potential applicability of our results in clinical practice. The selected liquid biopsy variables were
associated with standard commonly recognized end-points (PFS, OS, RR according to RECIST v1.1) and
afterwards integrated with clinico-pathological features able to affect the same endpoints.
Notable, even among patients achieving RR according to RECIST v1.1, liquid biopsy results were able to
further stratify patients and identify patients with differential duration of clinical benet, thus underlining
the limits of using RR as surrogate endpoint and as indicator for clinical communication and follow-up
planning during treatment.
Moreover, in our integrated model, we clearly demonstrated that longitudinal cfDNA analysis adds further
information when commonly used clinico-pathological predictive factors are available. In parallel, we
demonstrated the role of evaluating both cfDNA concentration and NGS results. In this context, we
previously demonstrated that cfDNA quantication can be considered a surrogate of tumor burden or
biological changes induced by treatment in a recent paper focussed in early identication with patients
potentially deriving “detrimental effects” from ICIs treatment12. On the other hand, we show here that the
two parameters are not fully overlapping and that NGS results, here evaluated by maxVAF values, are able
to further stratify risk of death and recurrence.

Page 15/20
The most important strengths of our study are the methodology in enrolment and the potential
applicability of the model. Notably, the choice of an early time-point (T2, three weeks after the start of
ICIs) permits to gure out potential applicability of the results on addressing early changes during rst-
line systemic treatment, with the nal aim to maximize treatment ecacy and avoid to administer more
toxic combination strategies when not needed.
On the other hand, the most important limitation of our study is the lack of a validation set of analyses to
conrm the results. However, we performed a cross-validation within the dataset to assess the robustness
of the optimal cut-points and we estimated the adjusted HR for the cut- points using a bootstrap
shrinkage procedure to reduce overtting.
In addition, further predictive information could be provided by including in the model qualitative
information about genetic characterization, in order to test potential inuence of specic genetic
alterations potentially able to affect the response to immunotherapy, such as STK11, KEAP1 and KRAS
mutation and further analyses are ongoing 29–32 but we believe that starting from information commonly
available in clinical practice might help the applicability of the model.
In our series we also estimated our model in a more homogenous population, being patients expressing
PD-L1 >/=50% in tumor cells and treated in rst line with pembrolizumab. We thus conrmed that our
model might be used to customize treatment approach in non-oncogene addicted advanced NSCLC
patients, since different treatment options including ICI single-agent and different combination strategies
are available of under clinical investigation.
In conclusion, we demonstrated the role of early monitoring of cfDNA concentration and maxVAF
quantication in plasma in predicting outcome of patients treated with ICIs and our integrated model
warrant further validation as a tool to customize treatment in non-oncogene addicted NSCLC.
Abbreviations
NSCLC
Non-small cell lung cancer
ICI(s)
Immune Checkpoint Inhibitor(s)
cfDNA
Circulating cell-free DNA
NGS
Next Generation Sequencing
VAF
Variant Allele Frequency
maxVAF
maximum Variant Allele Frequency

Page 16/20
PFS
Progression Free Survival
OS
Overall Survival
RR
Radiological Response
CR
Complete Response
PR
Partial Response
SD
Stable Disease
PD
Progressive Disease
Declarations
Ethics approval and consent to participate
The ethics committee of Istituto Oncologico Veneto evaluated and approved the study design and
informed consent (2016/82, 12/12/2016). Written informed consent was obtained from all patients
before study entry. The study was conducted in accordance with the precepts of the Helsinki declaration.
Consent for publication
The property of data belongs to the sponsor Istituto Oncologico Veneto – IOV and the authors have the
right to publish the data.
Availability of data and material
Molecular dataset is summarized in supplementary material and raw data are available upon reasonable
request.
Competing interests
The authors declare no competing interests.
Funding
This work was funded by IOV intramural research grant 2017–5×1000 (MAGIC-2, to S. Indraccolo and L.
Bonanno) and Ricerca corrente fundings from the Italian Ministry of Health.
Author’s contribution

Page 17/20
ABB: methodology, validation, investigation, writing—original draft, writing— review and editing and
visualisation. DBP: methodology, validation, formal analysis, data curation, writing—original draft, writing
—review and editing. ZE: methodology, investigation. IA: resources, data curation, editing. PA: resources,
data curation. CJ: investigation. ML: investigation. VG: resources. IS: conceptualisation, methodology,
writing—review and editing, supervision, project administration and funding acquisition. BL:
conceptualisation, methodology, data curation, writing—original draft, writing—review and editing,
visualisation, resources, supervision, project administration and funding acquisition.
References
1. Hendriks, L. E.
et al.
Non-oncogene-addicted metastatic non-small-cell lung cancer: ESMO Clinical
Practice Guideline for diagnosis, treatment and follow-up. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol.
34, 358–376 (2023).
2. Govindan, R.
et al.
Society for Immunotherapy of Cancer (SITC) clinical practice guideline on
immunotherapy for the treatment of lung cancer and mesothelioma. J. Immunother. Cancer 10,
e003956 (2022).
3. Reck, M.
et al.
Five-Year Outcomes With Pembrolizumab Versus Chemotherapy for Metastatic Non-
Small-Cell Lung Cancer With PD-L1 Tumor Proportion Score ≥ 50. J. Clin. Oncol. Off. J. Am. Soc.
Clin. Oncol. 39, 2339–2349 (2021).
4. Gandhi, L.
et al.
Pembrolizumab plus Chemotherapy in Metastatic Non-Small-Cell Lung Cancer. N.
Engl. J. Med. 378, 2078–2092 (2018).
5. Mok, T. S. K.
et al.
Pembrolizumab versus chemotherapy for previously untreated, PD-L1-expressing,
locally advanced or metastatic non-small-cell lung cancer (KEYNOTE-042): a randomised, open-label,
controlled, phase 3 trial. Lancet Lond. Engl. 393, 1819–1830 (2019).
. Paz-Ares, L.
et al.
First-line nivolumab plus ipilimumab combined with two cycles of chemotherapy in
patients with non-small-cell lung cancer (CheckMate 9LA): an international, randomised, open-label,
phase 3 trial. Lancet Oncol. 22, 198–211 (2021).
7. Herbst, R. S.
et al.
Atezolizumab for First-Line Treatment of PD-L1-Selected Patients with NSCLC. N.
Engl. J. Med. 383, 1328–1339 (2020).
. Sezer, A.
et al.
Cemiplimab monotherapy for rst-line treatment of advanced non-small-cell lung
cancer with PD-L1 of at least 50%: a multicentre, open-label, global, phase 3, randomised, controlled
trial. Lancet Lond. Engl. 397, 592–604 (2021).
9. Paz-Ares, L.
et al.
Pembrolizumab plus Chemotherapy for Squamous Non-Small-Cell Lung Cancer. N.
Engl. J. Med. 379, 2040–2051 (2018).
10. Ferrara, R.
et al.
Hyperprogressive Disease in Patients With Advanced Non-Small Cell Lung Cancer
Treated With PD-1/PD-L1 Inhibitors or With Single-Agent Chemotherapy. JAMA Oncol. 4, 1543–1552
(2018).

Page 18/20
11. Peters, S.
et al.
OA03.05 Analysis of Early Survival in Patients with Advanced Non-Squamous NSCLC
Treated with Nivolumab vs Docetaxel in CheckMate 057. J. Thorac. Oncol. 12, S253 (2017).
12. Zulato, E.
et al.
Longitudinal liquid biopsy anticipates hyperprogression and early death in advanced
non-small cell lung cancer patients treated with immune checkpoint inhibitors. Br. J. Cancer 127,
2034–2042 (2022).
13. Hegde, P. S., Karanikas, V. & Evers, S. The Where, the When, and the How of Immune Monitoring for
Cancer Immunotherapies in the Era of Checkpoint Inhibition. Clin. Cancer Res. Off. J. Am. Assoc.
Cancer Res. 22, 1865–1874 (2016).
14. Waldman, A. D., Fritz, J. M. & Lenardo, M. J. A guide to cancer immunotherapy: from T cell basic
science to clinical practice. Nat. Rev. Immunol. 20, 651–668 (2020).
15. Zulato, E.
et al.
Early assessment of KRAS mutation in cfDNA correlates with risk of progression and
death in advanced non-small-cell lung cancer. Br. J. Cancer 123, 81–91 (2020).
1. Bonanno, L.
et al.
Liquid biopsy and non-small cell lung cancer: are we looking at the tip of the
iceberg? Br. J. Cancer 127, 383–393 (2022).
17. Richards, S.
et al.
Standards and Guidelines for the Interpretation of Sequence Variants: A Joint
Consensus Recommendation of the American College of Medical Genetics and Genomics and the
Association for Molecular Pathology. Genet. Med. Off. J. Am. Coll. Med. Genet. 17, 405–424 (2015).
1. Eisenhauer, E. A.
et al.
New response evaluation criteria in solid tumours: Revised RECIST guideline
(version 1.1). Eur. J. Cancer 45, 228–247 (2009).
19. Champiat, S.
et al.
Hyperprogressive Disease Is a New Pattern of Progression in Cancer Patients
Treated by Anti-PD-1/PD-L1. Clin. Cancer Res. 23, 1920–1928 (2017).
20. Merker, J. D.
et al.
Circulating Tumor DNA Analysis in Patients With Cancer: American Society of
Clinical Oncology and College of American Pathologists Joint Review. J. Clin. Oncol. Off. J. Am. Soc.
Clin. Oncol. 36, 1631–1641 (2018).
21. Siravegna, G., Marsoni, S., Siena, S. & Bardelli, A. Integrating liquid biopsies into the management of
cancer. Nat. Rev. Clin. Oncol. 14, 531–548 (2017).
22. Wan, J. C. M.
et al.
Liquid biopsies come of age: towards implementation of circulating tumour DNA.
Nat. Rev. Cancer 17, 223–238 (2017).
23. Sivapalan, L.
et al.
Liquid biopsy approaches to capture tumor evolution and clinical outcomes
during cancer immunotherapy. J. Immunother. Cancer 11, e005924 (2023).
24. Cabel, L.
et al.
Circulating tumor DNA changes for early monitoring of anti-PD1 immunotherapy: a
proof-of-concept study. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 28, 1996–2001 (2017).
25. Raja, R.
et al.
Early Reduction in ctDNA Predicts Survival in Patients with Lung and Bladder Cancer
Treated with Durvalumab. Clin. Cancer Res. Off. J. Am. Assoc. Cancer Res. 24, 6212–6222 (2018).
2. Hwang, M.
et al.
Peripheral blood immune cell dynamics reect antitumor immune responses and
predict clinical response to immunotherapy. J. Immunother. Cancer 10, e004688 (2022).

Page 19/20
27. Assaf, Z. J. F.
et al.
A longitudinal circulating tumor DNA-based model associated with survival in
metastatic non-small-cell lung cancer. Nat. Med. 29, 859–868 (2023).
2. Vega, D. M.
et al.
Changes in Circulating Tumor DNA Reect Clinical Benet Across Multiple Studies
of Patients With Non–Small-Cell Lung Cancer Treated With Immune Checkpoint Inhibitors. JCO
Precis. Oncol. e2100372 (2022) doi:10.1200/PO.21.00372.
29. Pavan, A.
et al.
Role of next generation sequencing-based liquid biopsy in advanced non-small cell
lung cancer patients treated with immune checkpoint inhibitors: impact of STK11, KRAS and TP53
mutations and co-mutations on outcome. Transl. Lung Cancer Res. 10, 202–220 (2021).
30. Lim, T. K. H.
et al.
KRAS G12C in advanced NSCLC: Prevalence, co-mutations, and testing. Lung
Cancer 184, 107293 (2023).
31. Skoulidis, F.
et al.
STK11/LKB1 Mutations and PD-1 Inhibitor Resistance in KRAS-Mutant Lung
Adenocarcinoma. Cancer Discov. 8, 822–835 (2018).
32. Ricciuti, B.
et al.
Diminished Ecacy of Programmed Death-(Ligand)1 Inhibition in STK11- and
KEAP1-Mutant Lung Adenocarcinoma Is Affected by KRAS Mutation Status. J. Thorac. Oncol. 17,
399–410 (2022).
Figures
Figure 1
Kaplan-Meier curves for progression-free survival (A) and overall survival (B) of the study population
stratied into three risk groups based on their individual score in the nomogram (Supplementary Figure 1
and Supplementary Figure 2). Factors affecting PFS in multiple regression models were: cfDNA
concentration in plasma at baseline, maxVAF at T2, increase (T2-T1) of cfDNA, PD-L1 expression. Factors
affecting OS in multiple regression model were: cfDNA at baseline, increase (T2-T1) of cfDNA, maxVAF at
T2, histology and PD-L1 expression.

Page 20/20
Figure 2
Kaplan-Meier curves for progression-free survival (A) and overall survival (B) of patients treated with rst-
line pembrolizumab and stratied into three risk groups based on their individual score in the nomogram.
Factors affecting PFS in multiple regression models were: maxVAF at T2 and increase (T2-T1) of cfDNA.
Factors affecting OS in multiple regression model were: maxVAF at T2 and increase (T2-T1) of cfDNA.
Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.
SupplementaryTables20novwithcapstions.docx
Supplementarygurespluscaptions.docx