Impact of natremia on metastatic non small cell lung cancer patients receiving immune checkpoint inhibitors

Abstract

Hyponatremia has been established as a prognostic indicator of survival in metastatic non-small cell lung cancer (mNSCLC). Conversely, the influence of normal sodium levels remains unexplored. This study aims to investigate the impact of natremia in mNSCLC patients undergoing treatment with immune checkpoint inhibitors (ICIs). Clinical and biochemical data of patients treated with ICIs for mNSCLC were obtained. Availability of baseline sodium values was a study inclusion criterion. Patients were categorized into two groups based on the cut off sodium value, determined using the receiver operating characteristic curve. Subsequently, the influence of sodium levels on response rate (RR), disease control rate (DCR), progression-free survival (PFS), and overall survival (OS) was analyzed. PFS and OS were assessed via the Kaplan–Meier method. Univariate and multivariate Cox regression analyses were conducted to evaluate prognostic factors for PFS and OS. The analysis included 88 patients, of whom 73.1% were men, with a median age of 71 years (range, 47–91). A comparison between patients with baseline natremia ≥ 140 mEq/L (n = 43) and those with < 140 mEq/L (n = 45) revealed PFS durations of 7.0 vs. 2.1 months (p < .01) and OS durations of 15.6 vs. 6.8 months, respectively (p = .02). In the univariate survival analysis, pre-ICI serum sodium ≥ 140 mEq/L (p = .01) was associated with improved PFS, while factors associated with OS included brain metastasis (p = .05) and pre-ICI serum sodium ≥ 140 mEq/L (p = .02). In the multivariate analysis, pre-ICI serum sodium ≥ 140 mEq/L maintained a statistically significant association with OS (p = .04)..This study represents the first investigation into the impact of normonatremia in mNSCLC. Our findings suggest that serum sodium levels < 140 mEq/L at baseline and initial assessment are independently associated with poorer PFS and OS in mNSCLC patients undergoing first-line treatment with ICIs.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-024-81458-z.

Introduction

Non-small cell lung cancer (NSCLC) is a leading cause of cancer mortality worldwide, with a poor prognosis despite advances in treatment strategies¹. Immune checkpoint inhibitors (ICIs) have become a standard of care for patients with metastatic NSCLC. Immunotherapy can be used alone in first-line treatment or in combination with chemotherapy according to programmed cell death-ligand 1 (PD-L1) expression². However, despite encouraging results, only 20% of patients with NSCLC treated with immunotherapy demonstrate a durable response³.

Various predictive factors have been examined to uncover a molecular basis for the differing efficacy, revealing the significance of PD-L1 expression and tumor mutation burden^4,5. However it is probable that a complex interplay of genetic and acquired factors determines the effectiveness of immunotherapy in each patient⁶.

Hyponatremia is the most common electrolyte disorder in cancer patients, with low serum sodium levels linked to poor clinical outcomes in several type of cancer including NSCLC^7,8.

Retrospective experiences have highlighted the independent prognostic role of blood sodium level in NSCLC treated with first-line chemotherapy or target therapy⁹. Similarly, other retrospective analyses have demonstrated that hyponatremia is associated with a poorer prognosis in patients with NSCLC treated with immunotherapy, including shorter progression-free survival (PFS) and overall survival (OS)^10,11. Although the underlying mechanism is not clear-cut, hyponatremia in cancer often reflects a more severe disease state, complications from cancer treatment, and disrupted physiological balance (e.g., decreased renal function, poor nutritional status, cachexia), all of which are associated with worse survival outcomes.

The exact mechanisms leading to hyponatremia in NSCLC patients remain unclear. Syndrome of Inappropriate Anti-Diuresis (SIAD) represent 2–4% of hyponatremia, defined as sodium level < 135 mmol/L, in NSCLC¹². The presence of brain metastases and concomitant treatment with mannitol as well as the use of anti-epileptic drugs, characterized by impaired (i.e., reduced) water loss, could be cause of SIAD in NSCLC patients. Moreover, the presence of comorbidities, including heart or kidney failure, gastrointestinal leakage and cancer treatment, or concomitant medications as per opioid drugs, may contribute to the sodium loss.

Dysregulation of sodium homeostasis may affect immune cell function and cytokine signaling, thereby altering the efficacy of immunotherapy¹³. However, the relation between sodium levels and inflammation/immune response is bidirectional, since inflammation per se may modify sodium levels and induce hyponatremia (e.g., through secretion of IL-6)^14,15. Therefore, monitoring electrolyte levels is critical not only to identify patients at risk for adverse effects but also to optimize therapeutic strategies.

However, even though serum sodium levels are regularly measured at the start and throughout cancer treatment, the significance of natremia in NSCLC patients receiving ICIs has not been extensively investigated. This study aimed to evaluate the relationship between baseline serum sodium concentration and clinical outcomes in patients with metastatic NSCLC treated with immunotherapy.

Materials and methods

Patients and treatment

A retrospective analysis was conducted on clinical data from a consecutive series of patients diagnosed with metastatic NSCLC who received pembrolizumab or atezolizumab as their initial treatment regimen. The analysis covered the timeframe from March 2018 to March 2024 and was carried out at the Medical Oncology Center of Florence. Inclusion criteria for the study required availability of baseline serum sodium measurements taken within 10 days of treatment initiation. Various demographic and clinical parameters, including histological subtype, Eastern Cooperative Oncology Group (ECOG) performance status, choice of first-line therapy, and serum sodium levels, were systematically recorded for all patients. Pembrolizumab was administered intravenously at a fixed dose of 200 mg every three weeks according to local clinical protocols, continuing until disease progression or unacceptable toxicity occurred. Atezolizumab was administered intravenously at a fixed dose of 840 mg every two weeks under similar conditions^16,17. This study was conducted in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. All methods were performed in accordance with the relevant guidelines and regulations. The study protocol was approved by the Ethics Regional Ethical Committee of Tuscany, with registration number 17332_os, and written informed consent was obtained from all participants.

Treatment evaluation

Serum sodium levels were regularly assessed through a blood venous sample as part of laboratory evaluations. Treatment response evaluation occurred at three-month intervals through computed tomography scans, utilizing the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 for assessment ¹⁸. Treatment efficacy was gauged in terms of both OS and PFS. Any adverse events (AEs) arising during pembrolizumab or atezolizumab administration were closely monitored by investigators and promptly reported in line with the Common Terminology Criteria of Adverse Events (CTCAE) version 5.0¹⁹.

Efficacy outcomes

The primary aim of this study was to investigate the relationship between serum sodium levels and the efficacy of treatment and survival outcomes in non-small cell lung cancer (NSCLC) patients receiving immune checkpoint inhibitors (ICIs) as their initial therapeutic intervention. To accomplish this objective, patients were divided into two groups based on their serum sodium concentration. The main endpoints included PFS, defined as the time from the commencement of treatment to either disease progression or death, and OS, defined as the duration from the initiation of treatment to death from any cause. Additionally, secondary outcome measures included the disease control rate (DCR), which was defined as the proportion of patients achieving complete response (CR), partial response (PR), or stable disease (SD), as well as the objective response rate (ORR), indicating the proportion of patients achieving CR or PR ²⁰.

Statistical analysis

A thorough analysis was conducted to investigate potential prognostic associations, considering variables such as age, gender, histological subtype, smoking history, performance status, presence of brain metastasis, and serum sodium levels before commencement of systemic treatment. Continuous variables were presented as median values, along with ranges indicating the minimum and maximum values, while categorical variables were expressed as numerical counts and corresponding percentages. The cutoff value for serum sodium, derived from the population, was determined using receiver operating characteristic (ROC) curve analysis. PPFS and OS were estimated using the Kaplan-Meier method, and group comparisons were conducted using the log-rank test. Hazard ratios (HRs) and their associated two-sided 95% confidence intervals (CI) were calculated using the Cox proportional hazard model. Prospective prognostic factors for PFS and OS were initially evaluated through univariate analysis, and subsequently in multivariate analysis ²¹. P-value < 0.05 has been define as a cut-off that indicates statistical significance. For the examination of secondary outcomes, variables were dichotomized (adults aged 75 and older are generally referred to as older people²²), and Fisher’s exact test was utilized to establish correlations between dichotomized serum sodium values and clinical as well as biochemical parameters. Statistical analyses were conducted using MedCalc software version 14.

Results

Patient characteristics

A total of 88 patients met the eligibility criteria and were consequently included in the study, with their baseline characteristics succinctly outlined in Table 1. The median age of the patients was 71 years, ranging from 47 to 91 years. Among them, 57 (64.8%) were male, and 66 (81.5%) exhibited adenocarcinoma histology. Most patients (86.3%) presented with an ECOG-PS of ≤ 1, and 29 patients were identified as current smokers. Brain metastases were detected in 24 patients. Regarding treatment allocation, pembrolizumab was administered as a first-line treatment in 69.3% of cases, while atezolizumab was utilized in 30.7%. Serum sodium levels ranged from 133 to 146 mEq/L with a median value of 139 mEq/L. Notably, only one patient had values below the normal range (133 mEq/L) as well as only one above the range (146 mEq/L). Further details concerning the baseline characteristics of patients categorized based on their median serum sodium levels (< 140 or ≥ 140 mEq/L) are provided in Table 1. Importantly, no statistically significant differences were observed concerning demographic and clinical features between patients with serum sodium levels < 140 mEq/L and those with levels ≥ 140 mEq/L during the baseline evaluation. Demographic and clinical features were analyzed after disaggregating the data by gender. No statistically significant differences emerged between patients with serum sodium levels < 140 mEq/L and those with levels ≥ 140 mEq/L during the baseline evaluation in both the male and female groups (Table S1, S2).

Table 1.

Patients’ baseline characteristics.

	All patients (n = 88)	Na < 140 mEq/L (n = 45)	Na ≥ 140 mEq/L (n = 43)	p
Age†
Years	73 (47–91)	72 (47–91)	73 (52–89)	0.80
Gender§
Male	57 (64.8)	33 (73.3)	24 (55.8)	0.11
Histology§
Adenocarcinoma	66 (81.5)	32 (80)	34 (82.9)	0.82
ECOG-PS§
2	12 (13.6)	8 (17.8)	4 (9.3)	0.31
Smoker*
Current	29 (34.9)	13 (30.9)	16 (39.0)	0.40
Brain metastases§
Yes	24 (27.3)	15 (33.3)	9 (20.9)	0.21
Bone metastases
Yes
First-Line Therapy§
Pembrolizumab	61 (69.3)	34 (75.6)	27 (62.8)	0.22
Atezolizumab	27 (30.7)	11 (24.4)	16 (37.2)
PD-L1 expression
> 80%	49 (55.7)	23 (51.1)	26 (60.5)	0.42
Baseline Na+
Median (range), mEq/L	139 (133–146)	138 (133–139)	142 (140–146)	< 0.001

PS: performance status; ECOG; Eastern Cooperative Oncology Group: NA: serum sodium. * Data from 81 patients. Data are expressed as median (range)^†or absolute number (percentage)^§.

Efficacy outcomes and best responses

The assessment of efficacy outcomes and the determination of the best response were based on the cut-off for serum sodium levels established by the ROC curve (≥ 140 mEq/L). Notably, the PFS was notably prolonged in the cohort characterized by higher baseline sodium levels (≥ 140 mEq/L) compared to the group with lower levels (< 140 mEq/L) (7.0 vs. 2.1 months, respectively; p < .01) (Fig. 1). Similarly, the median OS demonstrated an extension in the group with higher baseline sodium levels (≥ 140 mEq/L) in contrast to the cohort with lower levels (< 140 mEq/L) (15.6 vs. 6.8 months, respectively; p = .02) (Fig. 2). No significant differences were observed in the ORR and DCR between patients with serum sodium levels above or below 140 mEq/L during the baseline assessment (p = .07 and p = .1) as reported in Table 2. A subgroup analysis in patients with brain metastases (BMs) did not show a statistically significant difference between patients with higher baseline sodium levels and those with lower levels in terms of PFS (5.8 vs. 1.4 months, respectively; p = .06) and OS (8.0 vs. 2.1 months, respectively; p = .2) (Table S3) When considering data by gender, no differences in survival outcomes were observed among female patients with higher or lower sodium levels (PFS: 6.9 vs. 1.4 months, respectively, p = .08, and OS: 15.6 vs. 2.3 months, respectively, p = .12). However, in male patients, a sodium level ≥ 140 mEq/L was associated with better PFS compared to lower sodium levels (9.3 vs. 2.8 months, respectively; p = .02), but not for OS (15.0 vs. 7.3 months, respectively; p = .1) (Table S4). In the univariate survival analysis, a baseline serum sodium level ≥ 140 mEq/L was associated with improved PFS (HR 0.52, 95% CI 0.31–0.88, p = .01), while factors associated with OS included the presence of brain metastasis (HR 1.76, 95% CI 1.01–3.10, p = .05) and baseline serum sodium levels ≥ 140 mEq/L (HR 0.54, 95% CI 0.31–0.93, p = .02). In the multivariate analysis, baseline serum sodium levels ≥ 140 mEq/L (HR 0.48, 95% CI 0.27–0.86, p = .01), BMs (HR 2.26, 95% CI 1.24–4.10, p = .007) and brain metastases(HR 1.91, 95% CI 1.04–3.50, p = .03) maintained a statistically significant association with PFS; while sodium levels ≥ 140 mEq/L (HR 0.53, 95% CI 0.28–0.99, p = .04) and brain metastases (HR 2.30, 95% CI 1.20–4.41, p = .01) with OS. For comprehensive results of the univariate and multivariate analyses, please refer to Table 3.

Table 3.

Univariate analysis for PFS and OS.

	HR	CI 95%	p
Progression free Survival
Age
> 75	1.25	0.74–2.11	0.39
Gender
Male	0.92	0.55–1.54	0.76
Histology
Adenocarcinoma vs. other	0.88	0.45–1.70	0.70
ECOG
≥ 2	0.80	0.16–3.88	0.78
Brain metastases
Yes	1.54	0.90–2.65	0.11
Bone metastases
Yes	1.59	0.94–2.70	0.08
First-line therapy
Pembrolizumab vs. atezolizumab	1.37	0.74–2.55	0.30
Smoker
No former vs. current	1.10	0.65–1.85	0.71
PDL1
≥ 80%	0.79	0.48–1.31	0.37
Baseline Na+
Na ≥ 140 mEq/L	0.52	0.31–0.88	0.01
Overall Survival
Age
> 75	1.42	0.81–2.47	0.21
Gender
Male	0.98	0.56–1.72	0.96
Histology
Adenocarcinoma vs. other	0.97	0.47–2.02	0.95
ECOG
≥ 2	1.42	0.66–3.01	0.36
Brain metastases
Yes	1.76	1.01–3.10	0.05
Bone metastases
Yes	1.30	0.73–2.30	0.36
First-line therapy
Pembrolizumab vs. Atezolizumab	1.44	0.74–2.83	0.28
Baseline Na+
Na ≥ 140 mEq/L	0.54	0,31-0.93	0.02
Smoker
No former vs. current	1.05	0.60–1.86	0.84
PD-L1
≥ 80%	0.74	0.43–1.26	0.27

PFS: progression free survival; OS: overall survival; ECOG: Eastern Cooperative Oncology Group; Na: sodium; PD-L1: programmed death ligand 1.

Fig. 1.

Kaplan-Meier survival curve of progression free survival according to baseline sodium value.

Fig. 2.

Kaplan-Meier survival curve of overall survival according to baseline sodium value.

Table 2.

Best response, PFS and OS according to serum sodium values.

	RR n, (%)	DCR n, (%)	Median PFS months, (95% CI)	Median OS months, (95% CI)
All patients (n = 88)	30 (46.9)	48 (75.0)	5.3 (2.2–8.7)	7.8 (4.2–16.6)
Baseline Na+
≥ 140 mEq/L (n = 43)	19 (63.3)	26 (54.1)	7.0 (3.6–25.6)	15.6 (14.1–25.3)
< 140 mEq/L (n = 45)	11 (36.6) p = .07	22 (45.8) p = .1	2.1 (1.4–6.2) p < .01	6.8 (2.1–10.4) p = .02

RR: response rate; DCR: disease control rate; PFS: progression free survival; OS: overall survival; Na: serum sodium; CI: confidence interval; ICI: immune checkpoint inhibitor.

Table 3.

Multivariate analysis for PFS and OS.

	HR	CI 95%	p
Progression free Survival
Age
> 75 years	1.36	0.76–2.43	0.29
Gender
Male	0.81	0.46–1.42	0.47
Histology
Adenocarcinoma vs. other	0.85	0.41–1.73	0.65
ECOG
≥ 2	0.81	0.30–2.16	0.68
Brain metastases
Yes	1.91	1.04–3.50	0.03
Bone metastases
Yes	2.26	1.24–4.10	0.007
First-line therapy
Pembrolizumab vs. atezolizumab	1.26	0.62–2.56	0.51
Smoker
No former vs. current	1.13	0.87–2.77	0.13
PDL1
≥ 80%	0.87	0.51–1.49	0.62
Baseline Na+
Na ≥ 140 mEq/L	0.48	0.27–0.86	0.01
Overall Survival
Age
> 75 years	1.79	0.96–3.35	0.06
Gender
Male	0.83	0.44–1.55	0.56
Histology
Adenocarcinoma vs. other	0.92	0.42-2.00	0.84
ECOG
≥ 2	1.45	0.57–3.68	0.43
Brain metastases
Yes	2.30	1.20–4.41	0.01
Bone metastases
Yes	1.82	0.95–3.49	0.07
First-line therapy
Pembrolizumab vs. Atezolizumab	1.56	0.72–3.38	0.25
Baseline Na+
Na ≥ 140 mEq/L	0.53	0,28-0.99	0.04
Smoker
No former vs. current	1.42	0.75–2.68	0.27
PD-L1
≥ 80%	0.89	0.49–1.62	0.72

PFS: progression free survival; OS: overall survival; ECOG: Eastern Cooperative Oncology Group; Na: sodium; PD-L1: programmed death ligand 1. Cox regression model has been used.

Discussion

The treatment of NSCLC has seen significant advancements, particularly in the use of initial immunotherapy, which has contributed to improve patient prognosis². Pembrolizumab and atezolizumab represent the standard of care as first-line therapy, alone or combined with chemotherapy based on the PD-L1 expression. However, a significant portion of patients does not respond adequately to ICI therapy or experiences limited benefits³. At present, PD-L1 expressions and tumor mutation burden appear to hold the most promising potential as biological markers for predicting the response to immunotherapy. However, no single predictor is adequate to predict patient survival benefit, and much effort is directed at developing prognostic models that integrate multiple predictors ²³.

Hyponatremia is the most prevalent electrolyte imbalance in cancer patients, with low serum sodium levels associated with poor clinical outcomes in various cancers, including NSCLC^7,8. The precise mechanisms causing hyponatremia in NSCLC patients are not well understood, but different factors such as SIAD, comorbidities (e.g., heart or kidney failure, and gastrointestinal leakage), cancer treatments, and concomitant medications like opioids, may contribute to it¹².

Our investigation marks the inaugural attempt to evaluate the predictive significance of serum sodium levels in NSCLC patients undergoing first-line treatment with pembrolizumab or atezolizumab. We discovered a notable correlation between a baseline serum sodium level of ≥ 140 mEq/L and markedly improved PFS and OS (p < .01 and p = .2, respectively). These findings echo our previous research, where we observed that lower sodium levels (< 140 mEq/L) were linked to worse PFS and OS in metastatic renal cell carcinoma (mRCC) patients undergoing first-line tyrosine kinase inhibitor therapy or second-line nivolumab treatment. Similarly, in metastatic colorectal cancer patients treated with aflibercept and chemotherapy, baseline sodium levels ≥ 140 mEq/L were significantly associated with prolonged PFS and OS compared to lower sodium levels, indicating a potential prognostic role of baseline serum sodium levels in these patient cohorts.

Notably, Rinaldi et al., evaluated the prognostic impact of hyponatremia in NSCLC patients with BMs. They found that patients with BMs and hyponatremia had a mOS of 10.1 months versus 13.1 months for eunatremic patients with BMs, suggesting an important prognostic role of sodium level in the management these patients ²⁵. Our recent study found that elevated sodium levels (≥ 140 mEq/L) pre- and post-ICI treatment correlated with better survival outcomes in mRCC patients with BMs ²⁶. In this, no statistically significant differences were observed between patients with BMs and elevated sodium levels (≥ 140 mEq/L) compared to BMs a lower sodium levels (< 140 mEq/L) in term of OS (p = .20) and PFS (p = .06), respectively.

The importance of sex-specific factors in cancer management is increasingly recognized, making research into the influence of sex and gender on the disease more pressing than ever. Scientific evidence demonstrates variations in disease occurrence, as well as differences in treatment responses and adverse effects, between the two sexes. Examining the role of gender as distinct from sex in cancer immunotherapy is a complex issue due to the numerous confounding factors that are challenging to assess. n our analysis, we evaluated whether there were differences between males and females with sodium levels above and below the cutoff in terms of PFS and OS. We found a statistically significant difference only in the male group, where higher sodium levels correlated with better PFS (p = .02).

A recent post hoc analysis of two phase 3 clinical trials, IMmotion151 and IMvigor 211, revealed that elevated baseline sodium levels predict a favorable response to immunotherapy and improved outcomes in patients with mRCC and metastatic urothelial carcinoma undergoing immunotherapy. In contrast to other key serum electrolytes like potassium, magnesium, and calcium, only sodium displays a linear correlation with favorable prognosis during immunotherapy, indicating a potential beneficial role of increased sodium levels. Notably, after adjusting for prognostic factors, increased sodium levels did not confer a better prognosis in the comparator arms of the evaluated trials, sunitinib and chemotherapy, respectively, suggesting that the predictive value of baseline sodium may be specific to immunotherapy. Increasing evidence suggests a direct immune-modulating impact of sodium ²⁷. Preclinical models have demonstrated that a high salt diet can impede tumor growth in mice by bolstering anti-tumor immunity and modulating the differentiation of myeloid-derived suppressor cells ^28,29. Moreover, sodium and chloride, collectively referred to as salt, play a role in priming a distinct activation state of macrophages ³⁰.

Emerging in vivo findings suggest a role for sodium storage in immune system activation, with potential implications for cancer therapy ³¹. Specifically, sodium intake may influence the activation state of the immune system by directly affecting T helper cell subtypes and innate immune cells in various tissues. Additionally, elevated sodium consumption may alter the composition of the intestinal microbiota, indirectly impacting immune cells. These revelations hint at sodium’s regulatory functions in numerous health conditions, encompassing cardiovascular disease, inflammation, infection, autoimmunity, and possibly even cancer treatment.

Our current study, consistent with our prior findings ^32,33, suggests a plausible association between elevated sodium levels and enhanced response to ICI in metastatic NSCLC patients. These discoveries hold promise for refining the management of NSCLC patients.

However, our research possesses several limitations. Firstly, the choice of evaluating only serum sodium at baseline, indeed it should be considered that the variation in serum Na levels over time and its potential impact on clinical outcomes (PFS and OS). Secondly, the retrospective nature regarding a single center and the absence of a comprehensive evaluation of patient comorbidities and concurrent medications notably antihypertensive drugs. As a future perspective, it would be valuable to investigate these data through a gender lens, considering that women tend to use a higher number of medications such as carbamazepine, selective serotonin reuptake inhibitors, phenothiazines, MDMA, haloperidol, chlorpropamide, and diuretics (particularly thiazide diuretics) that are known to induce electrolyte imbalances, including hyponatremia compared to men ^34,35. Other factors such as the circumstances surrounding sample collection could influence sodium levels in these individuals. Finally, nevertheless that a model using fixed pre-planned variables might provide more reliable outcomes, we have opted for the stepwise method due to the exploratory nature of our study. Further prospective studies are warranted to validate our findings.

Conclusion

In summary, our investigation has unveiled that NSCLC patients treated with pembrolizumab or atezolizumab as first-line therapy, the presence of a baseline serum sodium level of ≥ 140 mEq/L is associated with extended PFS and OS in comparison to those with sodium levels < 140 mEq/L. Higher sodium levels may constitute a crucial factor linked to improved survival outcomes in NSCLC patients undergoing immunotherapy. This implies its potential inclusion as an additional parameter in patients’ risk assessments. Further investigations are warranted to validate our findings.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

Dr G.R. have full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design: G.R., M.C., L.A. Acquisition, analysis, or interpretation of data: S.M., E.C., F.M., B.N., S.S., C.M., M.G.a Michelet. Drafting of the manuscript: Martina Catalano, Sara Fancelli. Critical revision of the manuscript for important intellectual content: Giandomenico Roviello, Lorenzo Antonuzzo. Statistical analysis: Martina Catalano, Giandomenico Roviello. Supervision: Giandomenico Roviello, Lorenzo Antonuzzo, Serena Pillozzi

Funding

This study was conducted without support.

Data availability

The datasets used and/or analysed during the current study available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.