Role of Volumetric Modulated Arc radioTherapy (VMAT) in the adjuvant treatment of locally advanced differentiated thyroid cancer: single institution experience

Abstract

Purpose

The optimal management of locally advanced differentiated thyroid cancer (DTC) remains a topic of debate, and the adoption of adjuvant external beam radiotherapy (EBRT) is discussed. This study aimed to evaluate the efficacy of EBRT with volumetric modulated arc radiotherapy (VMAT) in improving locoregional relapse-free survival (LRFS) in addition to surgery and radioiodine therapy (RAI). Secondary objective was to assess the tolerability of EBRT with VMAT.

Methods

This monocentric, retrospective study included patients with locally advanced DTC, positive resection margins (R1) or lymph node metastases with extra-nodal extension who underwent total thyroidectomy and RAI. Patients were grouped based on whether they received adjuvant EBRT after surgery and RAI, and outcomes were compared.

Results

Overall, 25 patients were included. EBRT group comprised 10 patients and non-EBRT group 15 patients. In the EBRT group, none experienced locoregional recurrence, compared to 60% in the non-EBRT group. Comparison of the survival curves showed a statistically significant difference (p = 0.043). Additionally, no severe treatment-related adverse effects were reported, indicating good tolerability of EBRT. Distant Disease-Free Survival (DDFS) was similar between the groups.

Conclusion

Adjuvant EBRT using VMAT improves LRFS in patients with locally advanced DTC, without adding significant toxicity. Locoregional control potentially avoid repeated neck surgeries and their associated side effects. Despite some limitations, our findings support the integration of EBRT into treatment protocols for selected high-risk patients. Further studies are needed to confirm these results and to evaluate the long-term impact of EBRT on survival and quality of life.

Introduction

Differentiated thyroid carcinoma (DTC) is a widely prevalent condition. Despite its high incidence, the mortality rate remains low, with survival rates reaching 99%. However, a small percentage of these patients present at diagnosis with locally advanced disease, which has higher risk of recurrence and mortality, and requires multidisciplinary therapeutic evaluation [1].

While surgery, radioactive iodine ablation (RAI) and suppressive therapy with levothyroxine are regularly used tools in the treatment of high-risk DTC, the use of external beam radiotherapy (EBRT) remains debated. EBRT has been recognized as a treatment that improves locoregional control in DTC patients who present high-risk factors for locoregional recurrence, including those with pT4 tumors, positive surgical margins, or extensive extrathyroidal or extra-nodal extension. However, due to the retrospective nature of the available studies, differences in patient selection criteria, and contrasting conclusions, the role of EBRT in these patients is still controversial [2].

The proximity of the thyroid gland and cervical lymph node chains to critical structures such as the spinal cord, larynx, oropharynx, and esophagus has historically made it difficult to deliver adequate radiation doses to target volumes using older radiotherapy techniques. As a result, concerns about potential side effects have contributed to ongoing resistance toward the routine adoption of adjuvant radiotherapy in patients with an elevated risk of locoregional recurrence [3].

The recent improvement in radiotherapy delivery techniques, such as Intensity-Modulated Radiotherapy (IMRT) and Volumetric Modulated Arc Radiotherapy (VMAT), has made it possible to deliver more precise radiation, reducing the risk of toxicity [4–6].

The study aimed to contribute to the ongoing debate by sharing insights from our single-center experience.

The primary objective was to assess locoregional relapse-free survival (LRFS) in patients with locally advanced DTC who underwent surgery, RAI and EBRT with VMAT compared to patients with a similar risk who underwent only surgery and RAI. The secondary objective of the study was to evaluate the tolerability of EBRT with VMAT.

Materials and methods

Study design

This is a retrospective monocentric study conducted at the Fondazione Policlinico Universitario Agostino Gemelli IRCCS in Rome. Advanced thyroid carcinomas were defined according to the European Society of Endocrine Surgeons [7]. We retrospectively evaluated patients with locally advanced or regional advanced DTC discussed within the multidisciplinary tumor board for thyroid carcinoma at our institution between 2010 and 2022. Staging was updated according to the most recent guidelines [8]. Among these, patients who met the following criteria were selected: age over 18 years; histological confirmation of differentiated thyroid carcinoma (papillary or follicular); locally advanced disease with at least one or more of the following criteria: pT4a or pT4b, positive resection margins (R1) or lymph node metastases with extra-nodal extension. Exclusion criteria were: age under 18 years; macroscopic post-surgical residual disease; medullary, anaplastic, or poorly differentiated predominant component thyroid carcinoma; presence of distant metastases or lymph node metastases on post-radioiodine scintigraphy; history of previous radiation therapy to the neck or chest; history of second malignancies; pregnancy and breastfeeding. All patients provided informed consent for the treatment. The study was conducted in accordance with the Declaration of Helsinki; this retrospective study was conducted according to the principles of good clinical practice and in accordance with GDPR regulations for privacy.

All patients were operated on by an expert surgical team, ensuring consistency in the surgical approach and post-operative care. Additionally, they were monitored using a standardized follow-up protocol, which provided a consistent method of patient management. All patients had undergone total extracapsular thyroidectomy, with central and/or unilateral or bilateral lateral lymph node dissection basing on the extent of the disease. En block extended resection was achieved when feasible [1]. Titanium clips were used to delimitate the site of suspected microscopic residual disease (R1) at the time of surgery, as also suggested by recent ESES position statement on definition and management of advanced thyroid carcinoma (statement 18) [7]. All patients were subsequently treated with high-dose RAI following suspension of thyroxine therapy, and their post-treatment scintigraphy showed no lymph node or distant localizations. In case of EBRT indication, all patients underwent FDG-PET and distant metastases were excluded.

Data recorded were: age at diagnosis, gender, date of surgical intervention, histological examination results, TNM staging, ATA risk classification, type of local infiltration (neural, muscular, esophageal, airways), and date and dose of radioactive iodine therapy. In patients undergoing adjuvant EBRT, the following additional data were collected: start and end dates of radiation therapy, delivered volumes, grading and type of adverse effects. For follow-up, LRFS was calculated, defined as the time free from locoregional disease recurrence (in the thyroid lodge or neck lymph nodes) detected through imaging exams (ultrasound with cytological confirmation, scintigraphy, or FDG-PET). For patients who experienced locoregional recurrence, the number of subsequent neck surgeries and radioactive iodine treatments, along with the cumulative total dose administered for recurrence management, were also reported. DDFS was also reported, defined as the time free from the appearance of distant metastases. In cases of progression, the initiation of systemic therapy, if any, was also noted. Adverse effects (AEs) were classified according to the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 guidelines [9].

In 2017, the Italian Association of Radiotherapy and Clinical Oncology (AIRO) issued clinical recommendations for EBRT use in DTC providing clinical indications, clinical target volumes and doses [10]. All patients who received adjuvant EBRT in the present report were treated using VMAT technique and in accordance with such recommendations. A total dose of 66 Gy was administered to the high-risk volume (tumor bed or positive nodal stations with extracapsular extension), 60 Gy to the intermediate-risk volume (positive nodal stations without extracapsular extension), and 50 Gy to the low-risk volume (elective nodal stations). All doses were delivered using standard fractionation at 2 Gy per day.

Statistical analysis

The statistical analysis was conducted using the “Statistical Package for Social Science (SPSS)”, version 20.0. Continuous variables are expressed as means ± standard deviation, and categorical variables are represented as frequencies. The normal distribution of the data was verified using the Kolmogorov-Smirnov test. The appropriate statistical test, parametric and non-parametric (Student’s t-test or Mann Whitney test, Chi-square test or Fisher’s exact test), was used for the analysis of the results. The locoregional relapse-free survival (LRSF), the distant disease-free survival (DDFS), and the overall survival (OS), defined as the interval of time from the first surgical intervention to the detection of locoregional recurrence, distant metastasis, or cause-specific death, respectively, were analyzed using the Kaplan–Meier method and compared with the log-rank test. Results with p < 0.05 were considered statistically significant.

Results

Study population

A total of 25 patients who met the inclusion criteria were selected. Among them, 10 received adjuvant EBRT according to the AIRO recommendations, while the remaining 15, despite meeting the same criteria, did not undergo EBRT due to patient refusal or because they had been evaluated before the adoption of the 2017 guidelines (therefore receiving a different treatment protocol which did not include radiotherapy). Based on this, patients were divided into two groups: the EBRT treatment arm and the non-EBRT control arm.

Patients in the EBRT group received treatment within 6 months of surgery (mean 6.1 months, SD ± 0.92). The average follow-up time for EBRT patients was 3.5 years (41.5 months, SD ± 8), whereas the control group had an average follow-up of 8 years (95.6 months, median 90 months, SD ± 53 months). The difference in observation time between the groups is due to the study design, as the control group included patients evaluated before the adoption of the AIRO guidelines, as explained in the previous paragraph.

The two groups were mostly homogeneous, except for mean age, which was higher in the EBRT group (61.6 vs. 50.7, p = 0.030), likely because younger patients were more concerned about side effects and thus more likely to refuse treatment. This aspect is further explored in the discussion. A full report of the main patient characteristics is presented in Table 1 in which it can be observed that the p-values comparing the two groups are not statistically significant, indicating that the groups are homogeneous and making the comparison meaningful.

Table 1.

Clinical and histopathological features of patients included in the present report

Study population (n = 25)	Adjuvant EBRT (n = 10)	No EBRT (n = 15)	P-value
Gender, n (%)			0.667
F	6 (60%)	11 (73.3%)
M	4 (40%)	4 (26.7%)
Age at diagnosis mean ± sd	61.6 ± 10.5	50.7 ± 12.6	0.030
Maximum diameter T mean ± sd mm	28.0 ± 15.6	19.7 ± 9.1	0.146
Histotype, n (%)			1.000
Classic papillary	3 (30%)	5 (33.3%)
Aggressive papillary var.	6 (60%)	8 (53.3%)
Poorly differentiated < 5%	1 (10%)	1 (6.7%)
Follicular	0 (0%)	1 (6.7%)
TNM Staging, n (%)			0.057
Stage I	2 (20%)	8 (53.3%)
Stage II	0 (0%)	2 (13.3%)
Stage III	8 (80%)	5 (33.3%)
T, n (%)			0.345
T3	1 (10%)	5 (33.3%)
T4	9 (90%)	10 (66.7%)
N, n (%)			0.102
N0	1 (10%)	3 (20%)
N1a	7 (70%)	4 (26.7%)
N1b	2 (20%)	8 (53.3%)
Type of infiltration, n (%)
Recurrent nerve	7 (70%)	6 (40%)	0.226
Muscle (macroscopic)	4 (40%)	8 (53.3%)	0.688
Esophagus	4 (40%)	1 (6.7%)	0.121
Trachea/larynx	4 (40%)	2 (13.3%)	0.175
Extranodal lymph nodes	3 (30%)	7 (46.7%)	0.678
Jugular	0 (0%)	2 (13.3%)	0.500

EBRT, External Beam Radiotherapy; F, female; M, male; TNM, Tumor-Nodes-Metastasis staging; T, tumor; N, node

Absence of recurrence in the EBRT treated group

None of the patients who underwent adjuvant EBRT developed locoregional recurrence throughout the entire follow-up period. By contrast, 9 patients of the control group (60%) developed locoregional recurrence. At 36 months (a shorter observation time than the median follow-up for EBRT patients), 4 of 15 untreated patients (26.6%) had already experienced locoregional recurrence. By five years post-diagnosis, the number of untreated patients with locoregional recurrence increased to 8 (53.3%). The average time to recurrence was 3.6 years (43 months). Figure 1 represents the LRFS curve for the two groups of patients until the end of the follow-up. The difference between the curves is evident and statistically confirmed by the log-rank test (p = 0.043). In the group treated with EBRT, LRFS was 100%, while in the non-EBRT group was 93.3% and 29.6% at 1 and 5 years respectively (Fig. 1).

Fig. 1.

LRFS curve in the EBRT group and in non-EBRT group. Green line = Adjuvant EBRT; Blue line = Non-EBRT. LRFS, Loco Regional Relapse-Free Survival; EBRT, External Beam Radiotherapy

Comparison of distant disease-free survival and overall survival showed no significant differences

DDFS curve was analyzed, with the onset of distant metastases diagnosed using total body scintigraphy with radioiodine or PET with 18-FDG. In the EBRT group, 2 patients (20%) developed distant metastases, while in the non-EBRT group, 6 patients (40%) had distant metastases. The following graph shows the survival curve for both groups. The difference was not statistically significant (p = 0.516), indicating similar rates of distant progression, at least in the first 60 months of follow-up (Fig. 2). Regarding OS, there were no deaths in the EBRT group, whereas there were 2 deaths (13%) in the non-EBRT group, occurring after an average of 16 years of disease.

Fig. 2.

DDFS curve in the EBRT group and in non-EBRT group. Green line = Adjuvant EBRT; Blue line = Non-EBRT; DDFS, Distant Disease-Free Survival; EBRT, External Beam Radiotherapy

Tolerability of VMAT

The acute adverse effects (AEs) experienced by patients treated with VMAT radiotherapy never reached grade 3, and all patients completed the treatment. One patient did not experience any side effects (10%), 5 patients (50%) had grade 1 effects, and 4 patients (40%) had grade 2 effects, of which 3 had epithelial desquamation, and 1 had dysphagia that required a semi-liquid diet. Regardless of the severity grade, 60% of AEs were skin toxicity: erythema (G1) or epithelial desquamation (G2); 60% had mucosal toxicity (all grade 1), and 30% had dysphagia, moderate (G2) in only one case. No other toxicities were detected. All AEs completely resolved in 100% of patients after treatment. To date, no long-term adverse effects have been observed.

Comparison of disease course between the two groups

Locoregional procedures. We compared the number of additional locoregional procedures and surgeries in the EBRT and non-EBRT group. The EBRT group, given the absence of recurrence, did not require additional procedures or surgeries. Considering the longer follow-up time available in the control group, we estimated the number of potentially avoidable procedures for the EBRT-treated patients, assuming the continued absence of recurrence. Patients in the non-EBRT group underwent an average of 0.8 surgeries per patient. Among those with a longer follow up (FUP), the average number of surgeries increased to 1.0 for patients with ≥ 3 years FUP and 1.8 for those with ≥ 10 years FUP. The total number of avoidable procedures (including surgery, radiometabolic therapy considered as a standalone event when administered as treatment and not as post-surgery ablation, and palliative EBRT) was 2 procedures per patient (median 2, mean 1.6, SD ± 1.4). In patients with longer FUP, this increased to an average of 2.4 and 3.4 procedures for patients with ≥ 3 and ≥ 10 years of FUP, respectively.

Systemic therapy. None of EBRT-treated patients required systemic therapy, whereas 3 patients (20%) in the non-EBRT group did. Additionally, 2 patients (13%) in the non-EBRT group eventually received radiotherapy for locoregional recurrences, occurring 9 and 12 years after the initial surgery.

Radioiodine therapy and cumulative doses. The average radioactive iodine dose administered to patients in the EBRT group, who were only subjected to radiometabolic therapy once, was 150 mCi (median, mean 143 mCi, SD ± 49). In contrast, the non-EBRT group underwent an average of 2 radiometabolic treatments with a cumulative mean dose of 241 mCi. Specifically, cumulative doses were higher in patients with longer FUP: considering only patients with FUP greater than 3 years, the average number of RAI treatments was 2.5, and average cumulative dose was 288 mCi (10656 MBq). In patients with a FUP greater than 10 years, the average number of RAI treatments was 3.2, and average cumulative dose was 352 mCi (13024 MBq).

Discussion

Our study aimed to assess the efficacy of adjuvant EBRT with VMAT following surgery and RAI in patients with pT4 tumors, positive surgical margins, or extensive extrathyroidal or extra-nodal extension. The role of adjuvant EBRT in DTC remains controversial, with inconsistent data in the literature contributing to the ongoing debate. By sharing our single-center experience, we sought to provide further insight into its potential benefits.

In 2016, the Endocrine Surgery Committee of the American Head and Neck Society issued four recommendations concerning the use of EBRT for locoregional control in DTC: (a) EBRT is advised for patients with gross residual or unresectable locoregional disease, excluding those under 45 years old with limited gross disease that is radioactive iodine-avid; (b) Routine use of EBRT as adjuvant therapy after complete resection of gross disease is not recommended; (c) EBRT may be considered post-resection for patients over 45 years old who have a high probability of microscopic residual disease and a low likelihood of responding to RAI; (d) Cervical lymph node involvement alone should not justify adjuvant EBRT [11]. More recently, an AIRO systematic review and meta-analysis supported the clinical recommendation to add EBRT to RAI after surgery for locally advanced DTC, though the strength of this recommendation was deemed conditional [12]. Due to the lack of strong evidence, adjuvant EBRT is not yet routinely considered in clinical practice [13]. In this context, a multidisciplinary approach is essential to personalize treatment decisions and carefully weigh the benefits and risks of adjuvant radiotherapy [14].

While multiple studies have demonstrated the effectiveness of adjuvant radiotherapy in improving locoregional control [15–17], its use remains controversial due to significant heterogeneity in study design and patient selection criteria. A 2020 meta-analysis compared the efficacy of adjuvant radiotherapy with non-irradiated control groups. Among the analyzed studies, only 9 of them were comparable, although varied widely in technique (3DCRT, IMRT, or VMAT), irradiated areas and doses [18]. Moreover, the limited number of patients, the difficulty in performing randomized trials and the different choices in inclusion criteria significantly limited the comparison of data. For instance, a prospective cohort study excluded high-risk patients, resulting in a low recurrence rate in the control group [19]. Another wide study involving 201 patients selected as advanced patients with lateral cervical lymph node involvement alone (N1b) without extra-nodal extension, contrarily to the fourth recommendation of American Head and Neck Society above mentioned [20]. Most literature combines indiscriminately patients with completely resected high-risk disease, microscopically positive margins, gross residual and locally recurrent tumors. Lastly, not all the studies performed RAI before adjuvant EBRT [21, 22].

In our study, we strictly adhered to the AIRO guidelines, selecting patients with pT4 DTC, positive resection margins, or metastases with extra-nodal extension, while excluding cases with gross residual disease, anaplastic carcinoma, or poorly differentiated histology. Additionally, treatment uniformity was ensured, as all patients were managed by the same multidisciplinary team, an important strength of our study. Despite the limitations due to the small sample size and the different observation time, our single-center experience demonstrated the high efficacy of adjuvant radiotherapy in reducing the rate of locoregional recurrence in patients with locally advanced thyroid carcinoma. Notably, none of the patients in the EBRT-group experienced recurrences, whereas the recurrence rate in the non-EBRT group was 60%. This finding remains significant despite the differences in observation time between the groups: at 36 months (shorter than the median follow-up for EBRT patients), 26.6% of non-irradiated patients had already developed locoregional recurrence. By five years post-diagnosis, over half of the untreated patients had experienced recurrence. Average time to recurrence was 3.6 years, which highlights the early risk of recurrence in this patient population. Indeed, DTC exhibits a higher recurrence rate within the first 5 years [23]. However, in intermediate and high-risk forms, recurrence remains a concern even after 10 or 20 years of follow-up [24]. Consequently, the design of prospective studies investigating high-risk DTC would necessitate extended observation periods, posing a significant challenge. Interestingly, a study with a prolonged observation time reported that the locoregional failure-free survival in the EBRT group remained statistically significant even 10 years after treatment [25].

Since almost all the studies were non-randomized and retrospective, patients who received radiotherapy tended to have more unfavorable baseline characteristics in terms of age, primary tumor stage, residual disease grade, and degree of infiltration [26, 27]. In our study, the comparative analysis of the two groups showed differences only in age of patients and size of the tumor. The mean age was significantly higher in the EBRT group (61.6 vs. 50.7, p 0.030), but the prognosis of these patients still proved to be better. The younger average age in the non-EBRT group likely resulted from the inclusion of patients who declined radiotherapy, often younger individuals concerned about potential side effects. Notably, other retrospective studies have showed a particular benefit of EBRT in older patients [28, 29]. Regarding tumor size, the EBRT group had a larger average maximum diameter compared to the non-EBRT group (28.0 vs. 19.7 mm), though this difference was not statistically significant. Interestingly, the EBRT group also included patients with sub-centimeter lesions, specifically one measuring 5 mm and another 6 mm. This highlights the importance of comprehensive preoperative imaging assessment, which assesses the risk of infiltration into adjacent structures and lymph node involvement, even in very small tumors. Current guidelines, in the absence of these additional risk factors, do not even recommend cytological examination for such small lesions [30].

The impact of adjuvant EBRT on OS remains unclear [31]. Most published studies have not reported a statistically significant difference in OS, and in some instances, OS has even been found to be worse [32]. The survival outcomes are influenced by the characteristics of the study population. Some studies included patients who received EBRT after multiple surgeries for recurrences, or those with distant metastases or who received chemotherapy treatment, which do not align with the correct definition of “adjuvant therapy” and were excluded from our study [33, 34]. Additionally, numerous studies considered patients with macroscopic residual disease post-surgery [35]. Our investigation did not provide enough observation time to adequately analyze overall survival.

The difference in follow-up duration between our two groups was due to the choice to include as control group patients taken in charge before the 2017 AIRO recommendations. This is a limit of our study, but on the other hand represents also a strength, as in the context of a slow-progressing disease with low mortality, the inclusion of patients with a long history of illness allowed us to monitor the number of treatments patients underwent over the years. Indeed, while locoregional recurrence is a negative prognostic factor, it is essential to note that it does not represent an irreversible situation, as recurrences can be successfully managed. For this reason, patients with recurrent disease often undergo repeated therapeutic interventions, including surgeries, RAI, minimally invasive ablations, and palliative radiotherapy. In our study, we observed that patients who did not receive adjuvant radiotherapy underwent repeated surgeries and locoregional procedures, and higher cumulative RAI doses. Additional procedures not only increase the risk of complications for the patient, but also have a greater impact on the healthcare system. Furthermore, locoregional recurrence can potentially carry an increased risk of anaplastic transformation or hematogenous dissemination, or directly affect quality of life due to the anatomical proximity to critical neck structures. Lastly, the potential economic impact and adverse effects of systemic drug therapies should also be considered [36]. For all these reasons, even in the absence of conclusive recommendations regarding improved OS and distant metastasis rates, reducing the risk of locoregional recurrence may itself be an important surrogate endpoint. Although therapies are available, preventing recurrences with adjuvant radiotherapy could be a more cost-effective strategy, particularly in low- and middle-income countries with limited resources and reduced patient adherence to follow-up.

The secondary objective of this study was to evaluate the tolerability of VMAT radiotherapy. Our findings indicated that the treatment was well tolerated, with no severe adverse effects. All patients completed the planned therapy, and no long-term effects were observed. These results are due to the advanced technology and equipment used at our center.

Existing literature, although limited, generally reports mild acute toxicity (grade 1 or 2), such as skin reactions and inflammation of the pharynx, esophagus, and salivary glands. Late toxicities were primarily low-grade xerostomia, with no severe chronic effects reported. Most studies have utilized IMRT or 3D CRT, showing a safer profile for the former, with fewer studies focusing on VMAT due to its more recent introduction [37]. No direct comparison between IMRT and VMAT for thyroid cancer is currently available. The variability in radiotherapy techniques, treatment volumes, and prescribed doses underscores the complexity of comparing different approaches, highlighting the need for individualized patient care.

The main limitations of this study stem from its retrospective design, small sample size, and limited observation period. These constraints were influenced by the study’s strict selection criteria and the decision to include in the irradiated group only patients treated after the 2017 AIRO guidelines, to ensure consistency in clinical indications, volume, dose planning and techniques. On the other hand, the longer observation period of the control group provided the opportunity to observe the long-term disease course. While these limitations pose challenges, they also represent strengths, as the extended follow-up of the control group enabled monitoring of recurrences and interventions over time. However, data with a longer observation period for treated patients will enhance the power of our results, and increasing the sample size could enable more detailed subgroup analyses based on age, histotype, lymph node involvement and molecular profiles.

Conclusions

Our experience confirms the effectiveness of adjuvant radiotherapy in improving locoregional control in patients with locally advanced differentiated thyroid cancer following surgery and radioactive iodine therapy (metabolic radiotherapy). In high-risk patients with locally advanced disease, positive resection margins with microscopic tumor residuals, and extensive structural invasion, EBRT can significantly reduce recurrence rates and potentially avoid repeated neck surgeries and their associated side effects. EBRT using VMAT has demonstrated a favorable toxicity profile, being well-tolerated without the occurrence of severe adverse effects. Longer follow up of this cohort and a larger sample size is required to enhance our results and to define the role of external beam radiotherapy in distant disease progression and cause-specific survival.

Author contributions

Conception and design of the study: C.M., B.F. and C.A.R. Acquisition, analysis, or interpretation of data: C.M., F.P., M.B., M.D.A. and A.L. Drafting the work and revising: C.M., B.F., L.Z., G.P. and G.S. Final approval of the article: E.D.R., M.A.G., M.R., L.T. and A.P.

Funding

This research received no external funding.

Data availability

The data presented in this study are available on request from the corresponding author due to privacy regulation.

Declarations

Conflicts of interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.

Institutional review board statement

The study was conducted in accordance with the Declaration of Helsinki, and ethical review and approval were provided (Prot. N. 0012205/15); this retrospective study was conducted according to the principles of good clinical practice and in accordance with GDPR regulations for privacy.

Informed consent

Informed consent was obtained from all subjects involved in the study.