We investigated efficacy, safety, and prognostic factors of reirradiation in patients with recurrent or second primary head and neck cancer.
Records of 75 consecutive patients treated with reirradiation between August 2005 and December 2013 were reviewed.
Median overall survival (OS) and cancer-specific survival (CSS) were 29.5 and 33.6 months. Median local control (LC) and progression-free survival (PFS) were 21.7 and 16.2 months. Univariate analysis showed that patients younger than 70 years, with a Karnofsky Performance Status (KPS) >90 or with 2 or less comorbidities at time of reirradiation, have a better OS; KPS >90 and biological equivalent dose (BED) >72 Gy positively influenced the PFS. At multivariate analysis, KPS at reirradiation was an independent predictive factor for OS, while BED was an independent predictive factor for CSS and OS. At univariate analysis, patients with planning target volume (PTV) >221 mL had worse LC and PFS rates, with results confirmed at multivariate analysis. The rate of fatal treatment-related adverse events was 6.7% (3 carotid blowout, 1 soft tissue necrosis, and 1 thromboembolic event).
This study confirms the role and outcomes of reirradiation. A careful selection of patients could minimize acute and late side effects and influence survival: elderly patients, with significant medical comorbidities or poor KPS, are worse candidate for reirradiation. Total dose delivered with reirradiation and PTV appear to be other potential prognostic factors. Further studies of dose escalation are needed to establish the total dose that could achieve better LC rates with a safer toxicity profile.
Tumori 2015; 101(5): 585 - 592
Article Type: ORIGINAL RESEARCH ARTICLE
AuthorsMichela Buglione, Marta Maddalo, Ercole Mazzeo, Pierluigi Bonomo, Luigi Spiazzi, Alessio Bruni, Fabiola Paiar, Luca Triggiani, Daniela Greto, Laura Rubino, Lorenzo Livi, Filippo Bertoni, Stefano Maria Magrini
- • Accepted on 13/05/2015
- • Available online on 15/06/2015
- • Published in print on 09/09/2015
This article is available as full text PDF.
The treatment of head and neck cancer (HNC) is mostly based on a multidisciplinary approach including surgery and radiochemotherapy (1, 2). Despite an increase in survival rates during the last few decades, due to advances in therapeutic procedures and treatment intensification, more than half of patients with locally advanced disease experience locoregional recurrence (3, 4).
Patients with early-stage HNC achieve high rates of long-term disease control, but could experience a second primary tumor arising within or in close proximity to previously irradiated tissues.
Evidence mostly based on retrospective series had established surgery as the standard of care for patients with recurrent disease (5-6-7-8-9); however, many patients are medically unfit for surgery, refuse surgery, or have unresectable disease. In these cases, 3 options can be discussed: supportive care only, palliative chemotherapy, or radiotherapy. As shown by several randomized trials, response rates after chemotherapy alone are poor, with median survivals not exceeding 6-8 months (10, 11). Therefore, radiation oncologists increasingly face the challenge of retreating patients who already experienced a high-dose radiotherapy course. Long-term survival rates have been reported after reirradiation of recurrent and second primary HNC in both prospective (12-13-14-15-16-17-18) and large retrospective series (19-20-21).
This retrospective study reports data in terms of efficacy and safety of a large population of patients with HNC treated with reirradiation. Endpoints of the analysis were overall survival (OS), cancer-specific survival (CSS), local control (LC), progression-free survival (PFS), and acute and late toxicities rates.
Materials and methods
We retrospectively reviewed medical records of patients with evidence of recurrence or second primary HNC, reirradiated at 3 Italian radiation oncology departments between August 2005 and December 2013.
Statistics were performed by SPSS software (SPSS Statistics 17.0; SPSS, Chicago, IL, USA).
A descriptive statistic of population characteristics and acute and late toxicity rates was carried out.
Toxicities were graded using the National Cancer Institute–Common Toxicity Criteria scale v 4.0 (22).
Kaplan-Meier survival analyses were used to estimate survival and local control rates.
The OS and CSS were calculated from the end of reirradiation up to the date of death or to the date of the last follow-up.
Local control was calculated from the end of reirradiation until the date of local recurrence (at the reirradiated site) or the date of the last follow-up.
The PFS was calculated from the end of reirradiation until the date of first evidence of disease or until the date of the last follow-up in case of disease-free patients.
Univariate analyses were performed using the log-rank test for all variables of interest: age at first diagnosis and at reirradiation (≤70 years versus >70 years), Karnofsky Performance Status (KPS) (23) at first diagnosis and at reirradiation (≥90 versus ≤80), Charlson Comorbidity Index (24, 25) (≤2 versus >2), disease presentation at reirradiation (recurrence versus second primary tumor), external beam radiotherapy (EBRT) technique, median biological equivalent dose (BED) delivered with reirradiation (<72 Gy versus ≥72 Gy), planning target volume (PTV) in quartiles and with a cutoff of 221 mL, concomitant chemotherapy, and LC (as a predictive variable for OS, CSS, and PFS).
Multivariate analyses were done using the Cox stepwise regression test. Variables used for these tests were the significant ones at the univariate analyses. All variables have been used as categorical. The BED was considered to possibly influence LC because it has shown a trend toward significance for this outcome at the univariate analysis. For this reason, it has been used for the respective multivariate analysis.
A p value <0.05 was considered statistically significant.
We identified a total of 75 adult patients, 61 male and 14 female. Fifty-nine patients had recurrent disease, while 16 patients presented a second primary tumor.
Patient and tumor characteristics
|KPS = Karnofsky Performance Status; RT = radiotherapy.|
|M/F, n (%)||61 (81.3)/14 (18.7)|
|Median (range) age, y|
|1st RT||58 (18-83)|
|KPS 1st RT, n (%)|
|KPS re-RT, n (%)|
|Charlson Comorbidity Index at re-RT, n (%)|
|1st RT disease site, n (%)|
|Oral cavity||6 (8)|
|Nasal cavity and paranasal sinuses||11 (14.7)|
|Salivary glands||1 (1.3)|
|Re-RT disease site, n (%)|
|Oral cavity||2 (2.7)|
|Nasal cavity and paranasal sinuses||11 (14.7)|
|Salivary glands||1 (1.3)|
|Exclusive neck nodal||19 (25.3)|
|Subcutaneous tissues||1 (1.3)|
Disease presentation and first radiotherapy course
Median age at first diagnosis was 58 years (range 18-83). All patients had excellent (90-100) or good (70-80) KPS and were treated between January 1971 and April 2013 with radical purpose. A total of 30.7% of patients received exclusive EBRT, 25.3% of patients were treated with concurrent chemoradiation, while 44% of patients had postoperative radiotherapy (29 with EBRT alone, 4 with concurrent chemotherapy). In all radiotherapy courses, the fractionation schedule was standard. The median dose was 66 Gy (range 40-70) in the curative setting and 60 Gy (range 50-70) in the postoperative one. The patient who received a curative dose of 40 Gy was treated at the beginning for a non-Hodgkin lymphoma, and then developed an oral cavity second primary tumor. The delivery of EBRT changed over the treatment period, ranging from a 2D or 3D conformal technique to intensity-modulated radiation therapy or tomotherapy. Concomitant chemotherapy was cisplatin in all 15 patients except 2 who received cetuximab.
Preretreatment clinical evaluation and reirradiation
Patients were treated with reirradiation between August 2005 and December 2013. Median time between the first radiotherapy course and reirradiation was 28 months (range 3-173) for recurrences and 144 months (range 35-446) for second primary tumors, respectively (p<0.001).
Median age at reirradiation was 64 years (range 25-87). The KPS was excellent in 47 patients, good in 27 patients, and poor in 1 patient.
At the time of preretreatment clinical evaluation, no severe (G3-G4) late dysphagia or dry mouth due to previous radiotherapy course was detected; only 2 patients had a G3 soft tissue fibrosis.
T and N stage at reirradiation
|Second primary tumor|
Fifty-two patients received EBRT, 12 patients were treated with concurrent chemoradiation (5 cisplatin and 7 cetuximab), and 11 patients had a postoperative radiotherapy (9 with EBRT alone, 2 with concurrent cetuximab). The adjuvant treatment was due to macroscopic persistence after surgery or to high-risk features at the anatomo-pathologic evaluation such as positive margins or massive lymph node involvement with extracapsular extension.
Twenty-six patients were treated with standard fractionated schedule, 30 patients underwent altered fractionation (24 with hyperfractionation once or twice a day and 6 with moderate once-a-day hypofractionation), and the remaining 19 patients received hypofractionated stereotactic body radiotherapy (SBRT) with CyberKnife. Patients who underwent SBRT were all treated in the same institution.
Radiotherapy technique in non-SBRT patients was 3D conformal radiotherapy in 9 patients, intensity-modulated radiotherapy in 9 patients, volumetric modulated arc therapy in 2 patients, and tomotherapy in 34 patients. In 2 patients with superficial soft tissue and nodal neck metastases, the technique used was an appositional direct electron field. In all patients, the target volume was represented by the gross tumor volume both for the primary tumor and the lymph nodes involved. Only 9 patients also received a prophylactic neck nodal irradiation.
In order to compare a heterogeneous set of fractionation schedules, a conversion of the total effective dose delivered to each patient to the BED was performed, using a α/β value of 10. The patient sample was split in 2 using the median BED as a cutoff (<72 Gy versus ≥72 Gy), in order to test its potential role in influencing outcomes. The BED in non-SBRT patients ranged from 29 to 84 Gy. In the SBRT group, the fractionation schedule was 25 Gy in 5 fractions in 5 patients (BED of 37.50 Gy), 30 Gy in 6 fractions in 11 patients (BED of 48 Gy), and 35 Gy in 7 fractions in 3 patients (BED of 59.50 Gy). All SBRT doses were prescribed at isodose 80%.
The median follow-up after reirradiation was 12 months (range 1-81).
Clinical results: OS and CSS
Median OS was 29.5 months; median CSS was 33.6 months. One- and 2- year OS rates were 64.4% and 58.2%, respectively. One- and 2- year CSS rates were 71.7% and 67.2%, respectively.
Twenty-six patients died of disease progression, 5 patients died of reirradiation adverse events, and 9 patients died of causes related neither to the treatment nor to the HNC.
Clinical outcomes with results of univariate analysis
|1 year OS||1 year CSS||1 year LC||1 year PFS|
|BED = biological equivalent dose; CSS = cancer-specific survival; EBRT = external beam radiotherapy; LC = local control; NS = not significant; OS = overall survival; PFS = progression-free survival; PTV = planning target volume; SBRT = stereotactic body radiotherapy.|
|Age at first diagnosis, y|
|≤70 (n = 60)||73.5||75.7||71.8||57.1|
|>70 (n = 15)||26.7||46.7||53.3||44.5|
|Age at reirradiation|
|≤70 years (n = 48)||75.4||78.2||71.6||59.2|
|>70 years (n = 27)||45.4||58.9||64||47|
|Karnofsky Performance Status at first diagnosis|
|≥90 (n = 65)||69.1||78||70.3||53.2|
|≤80 (n = 10)||38.1||72.6||44.4||65.5|
|Karnofsky Performance Status at reirradiation|
|≥90 (n = 47)||78.2||81||68.3||59|
|≤80 (n = 28)||41.9||53.6||67||44.7|
|Charlson Comorbidity Index|
|≤2 (n = 54)||69.9||75.2||63.2||56.1|
|>2 (n = 21)||48.4||58.6||84||49.6|
|Disease presentation at reirradiation|
|Recurrence (n = 59)||60.6||69.6||67||50.1|
|Second primary tumor (n = 16)||80||80||75||75|
|Non-SBRT (n = 56)||68.6||73.4||63.7||59.5|
|SBRT (n = 19)||52.6||66.1||81.8||42.9|
|<72 Gy (n = 38)||56.3||63.2||60.7||48.5|
|≥72 Gy (n = 37)||73.6||81.6||71.1||62.8|
|0-25th percentile (n = 13)||74.6||90.9||74.1||64.8|
|25th-50th percentile (n = 13)||39.5||47.9||82.1||28.8|
|50th-75th percentile (n = 13)||57.1||57.1||76.4||58.2|
|75th-100 percentile (n = 13)||41.1||44.9||0||0|
|<221 mL (n = 39)||57.4||66.1||76.8||51.3|
|≥221 mL (n = 13)||41.1||44.9||0||0|
|Concomitant chemotherapy during reirradiation|
|Yes (n = 15)||57.8||67.4||42.9||39.6|
|No (n = 60)||65.9||72.6||74.2||58.5|
|Yes (n = 42)||60.2||70.2||69.8|
|No (n = 33)||70.2||73.7||37.5|
Results of multivariate analyses
|Endpoint||Significant variables||Relative risk (confidence interval)||p Value|
|BED = biological equivalent dose; CSS = cancer-specific survival; KPS = Karnofsky Performance Status; LC = local control; OS = overall survival; PFS = progression-free survival; PTV = planning target volume.|
|OS||Age at diagnosis (≤70 years vs >70 years)||3.5 (1.6-7.7)||0.002|
|KPS at reirradiation (≥90 vs <80)||2.2 (1.1-4.4)||0.018|
|BED (≥72 Gy vs <72 Gy)||2.2 (1.1-4.7)||0.031|
|CSS||BED (≥72 Gy vs <72 Gy)||2.2 (1.0-4.9)||0.048|
|LC||PTV (<221 cm3 vs ≥221 cm3)||7.4 (2.3-23.5)||0.001|
|PFS||PTV (<221 cm3 vs ≥221 cm3)||3.1 (1.0-9.6)||0.045|
|Local control (yes vs no)||4.0 (1.8-9.2)||0.001|
Elderly patients (both at the time of first diagnosis and at the time of reirradiation) had a lower 1- and 2-year OS compared to those younger than 70 years (p = 0.002).
Patients with excellent KPS (≥90) at time of reirradiation had higher OS compared to those with a KPS ≤80 (p = 0.002).
At univariate analysis, having 2 or fewer major comorbidities (Charlson Comorbidity Index ≤2) before reirradiation is a prognostic factor for both a better 1-year OS (69.9% vs 48.4%, p = 0.039) and a better CSS (75.2% vs 58.6%, p = NS).
A trend for improved OS and CSS was seen with increasing radiation dose. Using a BED cutoff of 72 Gy (median BED), the 1-year OS and 1-year CSS were 56.3% and 63.2%, respectively, for patients who received less than 72 Gy (38 patients) and 73.6% and 81.6%, respectively, for those who received more than 72 Gy (37 patients), reaching statistical significance only for CSS (p = 0.043).
At multivariate analyses, KPS at reirradiation appears to be an independent variable that influences OS, while BED was an independent variable that influences both OS and CSS.
Patterns of failure: LC and PFS
Median LC was 21.7 months, while median PFS was 16.2 months. One- and 2-year LC rates were 68.6% and 46.4%, respectively. One- and 2-year PFS rates were 54.9% and 38.5%, respectively.
Among patients who died of disease progression (n = 26), 2 patients had progression at the site of reirradiation, 6 patients had systemic progression, and 18 patients had both local and systemic recurrence.
At the first clinical examination 3 months after the end of the treatment, 12, 20, and 27 patients had a response scored as <25%, between 25% and 75%, and ≥75%, respectively, while 11 patients had evidence of disease progression. In 5 patients, the clinical response cannot be assessed because they died before the first scheduled follow-up visit.
The PTV appears to be a critical aspect for both LC and PFS. A double univariate analysis was performed on this issue. The analysis per quartiles demonstrated that patients included in the upper quartile had worse LC and PFS outcomes compared to those included in all the other lower quartiles (p = 0.002 and 0.047, respectively). Based on this finding, we decided to carry on the same analysis using the 75th percentile (221 mL) as a cutoff for both the univariate and the multivariate analyses.
At multivariate analyses, the PTV (<221 mL vs ≥221 mL) is confirmed as an independent variable that influences both PFS and LC. The PFS was also influenced by LC.
Clinical results: toxicity
Acute and late toxicity (NCI-CTC scale v 4.0)
|NCI-CTC = National Cancer Institute–Common Toxicity Criteria.|
|Acute||Salivary duct inflammation||43||32||/||/|
|Superficial soft tissue fibrosis||36||23||2||/|
|Injury to carotid artery||57||/||1||3|
|Head and/or neck soft tissue necrosis||60||/||/||1|
|Osteonecrosis of jaw||60||1||/||/|
For 14 patients, data on late toxicity are not available: 8 patients died within 6 months from the end of reirradiation, 3 patients had an early disease progression and were submitted to palliative chemotherapy, and 3 patients were alive at the time of the analysis but were followed by other physicians.
Three patients died of a blowout of the carotid artery. One patient with a carotid hemorrhage underwent surgery both before reirradiation and before the first course of radiotherapy (lifetime dose of radiation 120 Gy). One patient was treated at presentation with postoperative radiotherapy and with concurrent radiochemotherapy at time of recurrence (lifetime dose of radiation 126 Gy). One patient aged 80 years at time of reirradiation received exclusive radiotherapy for both the first radiotherapy course and reirradiation (lifetime dose of radiation 120 Gy).
One patient died 7 months after the end of reirradiation with concurrent chemotherapy for a soft tissue necrosis in the oropharynx that probably brought on a carotid hemorrhage (lifetime dose of radiation 140 Gy).
One patient had a carotid stent after a transitory ischemic attack (classified as grade 3 injury to carotid artery) 4 months after SBRT treatment for a oropharynx recurrence; at the beginning, he was treated with postoperative concurrent chemoradiation.
One patient died 5 months after the end of palliative hypofractionated (30 Gy in 10 fractions) reirradiation for a second primary tumor of the oropharynx staged T2N3; the death was probably due to a thrombosis of the left jugular venous system, although the response after treatment was not assessable.
Patients with recurrent and second primary HNC have limited salvage options. For resectable disease, surgery has represented the standard of care; however, no randomized trials have been performed to compare surgery and reirradiation in this subset of patients.
Currently, there are no randomized trials that define the optimal approach and evidence concerning the role of reirradiation as a valuable option in these patients comes mainly from retrospective and phase II trials (12-13-14-15-16-17-18-19-20-21). Our study reports data of a large retrospective series with 1-year and 2-year survival rates comparable to or higher than those described in the literature. In 2007, long-term outcomes of RTOG 99-11 were reported by Langer et al (16). This was a prospective phase 2 trial on reirradiation with twice-daily radiotherapy plus chemotherapy with cisplatin and paclitaxel; 99 patients were enrolled, and the 1-year OS and PFS rates were 50.2% and 35%, respectively. In 2008, Spencer et al (18) reported a 1-year OS rate of 41.7% in another large phase 2 trial in which 81 patients were treated with twice-daily radiotherapy plus chemotherapy with hydroxyurea and 5-fluorouracil. Other prospective studies among smaller cohorts of patients evaluated the role of normo-fractionated radiotherapy (13, 14, 17) and SBRT (12, 15), with similar survival outcomes. None of these prospective and retrospective studies analyzes at the same time all the variables used for univariate and multivariate analyses in our series. A critical review of the published data shows that the selection of patients represents a crucial aspect in reirradiation; patients with significant medical comorbidities, poor performance status, or severe sequelae from prior radiation have typically been demonstrated to be poor candidates for reirradiation. The retrospective nature of this analysis did not allow us to report the features of patients who were excluded from reirradiation: the decision to treat or not to treat a patient was made by a multidisciplinary team at each institution and patients who did not undergo radiotherapy were treated with other single modality (surgery or chemotherapy) or addressed to supportive care. However, an analysis of our patients’ characteristics at the time of preretreatment clinical evaluation shows that no severe late dysphagia or dry mouth due to previous radiotherapy course was detected, and just 2 patients had a G3 soft tissue fibrosis: a selection of patients, based on late sequelae of the first radiotherapy treatment, has surely been done. In our analysis, KPS at time of reirradiation was an independent predictive variable for both OS and CSS. Tanvetyanon et al (26) published a paper that demonstrated the role of comorbidity and preexisting organ dysfunction as important prognostic factors for patients undergoing reirradiation. Looking at our results, we can draw similar conclusions, with poor OS and CSS rates in patients with a Charlson Index higher than 2 (2 or more major comorbidities). However, these findings have not been confirmed at multivariate analysis.
The optimum dose and fractionation schedule for recurrent HNC is currently under investigation.
In 2009, Heron et al (27) performed a dose-escalation phase I trial on 25 patients treated with SBRT in 5 dose tiers up to 44 Gy in 5 fractions. However, the goal of the study was to evaluate the safety and acute tolerability, and no conclusions were drawn on the efficacy of the higher dose regimen among the others. The same authors reported a retrospective analysis (28) on 96 patients showing that patients receiving more than 40 Gy in 5 fractions (BED equal to 72 Gy) had significantly higher LC compared with those receiving 36 Gy or lower. To our knowledge, no other studies of dose escalation have been published on patients reirradiated for a recurrence or a second primary head and neck tumor (and particularly on those treated with conventional techniques). In our series, with all the limitations due to the conversion of a SBRT high dose per fraction into a BED, an increasing BED seems to positively influence survival outcomes. In fact, even if the log-rank test on BED <72 Gy versus ≥72 Gy reached significance only for CSS, better rates with a BED ≥72 Gy were seen also for LC, PFS, and OS, and a larger sample of patients could possibly enhance these findings.
Ozyigit et al (29) reported a series of 51 patients treated for nasopharyngeal cancer with either SBRT or conventional radiotherapy. Our findings are similar to ones reported by these authors, with no benefit in terms of survival outcomes between patients treated with conventional and stereotactic techniques. To our knowledge, there are no other data in the literature that compare high dose per fraction SBRT with other EBRT techniques. It is important to note that in our non-SBRT population, we collected data of different techniques, including both less and more sophisticated ones: however, the patient sample for each single technique did not allow us to make a direct comparison between SBRT and standard fractionation therapy delivered with highly conformed techniques, such as tomotherapy, and further investigations are needed.
Treatment volume at time of reirradiation has been reported in few articles in the literature as a predictive variable of survival. Still, as shown in 2 recent reviews by Chen et al (30) and Cacicedo et al (31), the best volume cutoff has not yet been established. In fact, while De Crevoisier et al (21) used a treatment volume cutoff of 650 cm3, other authors (14, 30, 32) reported a cutoff of 25-27 cm3. For this reason, we decided to start our analysis without using a cutoff volume and we stratified patients in quartiles. This analysis allowed us to find a threshold at the 75th percentile (221 cm3), which correlates with LC and PFS outcomes.
Only 20% of our patients have been treated with concomitant chemotherapy and no conclusive statements can be done on the influence of concurrent chemotherapy on survival.
Our results on acute and late toxicities are similar to those reported in previous published experiences. As reported in the literature, the adverse event most feared in patients treated with reirradiation is the carotid blowout (33-34-35). In our experience, 3 patients (4%) died of a fatal hemorrhage and the overall rate of treatment-related death, which ranged in previous published series between 2.9% and 15.6% (16, 18, 19-20-21), was 6.7%. McDonald et al (33) report both surgery and chemotherapy as factors that may contribute to a carotid blowout: 4/6 patients who died of toxicity in our series had been treated with a more intensified multidisciplinary approach, including surgery and chemoradiation.
This study confirms the role and the outcomes of reirradiation in patients with a recurrence or a second primary HNC. A careful selection of patients could minimize acute and late side effects and probably influence survival: KPS, comorbidities, and type and number of previous interventions are major criteria that need to be taken into account. Fatal events seem to correlate more with treatments received before reirradiation than with dose fractionation or technology adopted: patients should be informed in detail of increased potential of side effects. Further studies are needed to establish the role of concurrent chemotherapy and the fractionation schedule and total dose delivered with reirradiation that could achieve better local control and safer toxicity profile. The PTV (and therefore, probably, recurrent/second primary tumor volume) is confirmed as an important variable that could predict LC and PFS.
- Buglione, Michela [PubMed] [Google Scholar] 1, * Corresponding Author (email@example.com)
- Maddalo, Marta [PubMed] [Google Scholar] 1
- Mazzeo, Ercole [PubMed] [Google Scholar] 2
- Bonomo, Pierluigi [PubMed] [Google Scholar] 3
- Spiazzi, Luigi [PubMed] [Google Scholar] 4
- Bruni, Alessio [PubMed] [Google Scholar] 2
- Paiar, Fabiola [PubMed] [Google Scholar] 3
- Triggiani, Luca [PubMed] [Google Scholar] 1
- Greto, Daniela [PubMed] [Google Scholar] 3
- Rubino, Laura [PubMed] [Google Scholar] 2
- Livi, Lorenzo [PubMed] [Google Scholar] 3
- Bertoni, Filippo [PubMed] [Google Scholar] 2
- Magrini, Stefano Maria [PubMed] [Google Scholar] 1
Radiation Oncology Department, University and Spedali Civili, Brescia - Italy
Radiation Oncology Department, Azienda Ospedaliero-Universitaria di Modena Policlinico, Modena - Italy
Radiation Oncology Department, Azienda Ospedaliero-Universitaria Careggi, Firenze - Italy
Medical Physics Department, Spedali Civili di Brescia, Brescia - Italy