Reirradiation of salivary gland tumors with carbon ion radiotherapy at CNAO

Radiotherapy and Oncology 145 (2020) 172–177

Contents lists available at ScienceDirect

Radiotherapy and Oncology journal homepage: www.thegreenjournal.com

Original Article

Reirradiation of salivary gland tumors with carbon ion radiotherapy at CNAO B. Vischioni a,⇑, B. Dhanireddy a,b, C. Severo a,c, M. Bonora a, S. Ronchi a, V. Vitolo a, M.R Fiore a, E. D’Ippolito a, R. Petrucci a, A. Barcellini a, E. Ciurlia a,d, A. Iannalfi a, A. Hasegawa a,e, S. Molinelli a,e, A. Mirandola a,e, F. Valvo a, R. Orecchia a,f a Radiation Oncology Clinical Department, National Center for Oncological Hadrontherapy (CNAO), Pavia, Italy; b Radiation Medicine, Albert B. Chandler Hospital, University of Kentucky, USA; c Section of Radiological Sciences, University of Messina, Messina; d Radiation Oncology Department, Vito Fazzi Hospital, Lecce, Italy; e Radiation Oncology Department, Osaka Heavy Ion Therapy Center, Osaka, Japan; f Department of Radiotherapy, European Institute of Oncology, Milan, Italy

a r t i c l e

i n f o

Article history: Received 1 August 2019 Received in revised form 6 January 2020 Accepted 7 January 2020

Keywords: Salivary gland tumors Carbon ion radiotherapy

a b s t r a c t Aims: To report oncologic and functional outcomes in terms of tumor control and toxicity of carbon ion radiotherapy (CIRT) in reirradiation setting for recurrent salivary gland tumors at CNAO. Methods: From November 2013 to September 2016, 51 consecutive patients with inoperable recurrent salivary gland tumors were retreated with CIRT in the frame of the phase II protocol CNAO S14/2012C for recurrent head and neck tumors. Results: Majority of pts (74.5%) had adenoid cystic carcinoma, mainly rcT4a (51%) and rcT4b (37%). Median dose of prior photon based radiotherapy was 60 Gy. Median dose of CIRT was 60 Gy [RBE] at a mean of 3 Gy [RBE] per fraction. During reirradiation, 19 patients (37.3%) experienced grade G1 toxicity, 19 pts (37.3%) had G2 and 2 pts (3.9%) had G3. Median follow up time was 19 months. Twenty one (41.2%) patients had stable disease and 30 (58.8%) tumor progression at the time of last follow up. Furthermore, 9 (18%) patients had G1 late toxicity, 19 (37%) had G2 and 9 (17. 5%) had G3. Using the Kaplan Meier method, progression free survival (actuarial) at one and two years were 71.7% and 52.2% respectively. Estimated overall survival (actuarial) at one and two years were 90.2% and 64%, respectively. Conclusions: CIRT is a good option for retreatment of inoperable recurrent salivary gland tumors with acceptable rates of acute and late toxicity. Longer follow up time is needed to assess the effectiveness of CIRT in reirradiation setting of salivary gland tumors. Ó 2020 Elsevier B.V. All rights reserved. Radiotherapy and Oncology 145 (2020) 172–177

Treatment for salivary gland tumors that are deemed potentially resectable, is radical surgical resection ensuring free margins, and almost always followed by post-operative radiotherapy (RT). In the absence of prospective randomized studies, a wide spread of retrospective literature has shown postoperative RT to significantly improve cure rates, especially for patients (pts) with incompletely resected tumors [1–3]. Despite combined modality approach for primary salivary gland tumors, the results have remained suboptimal, and many pts experience local failures in the long run [4–6]. For locally recurrent tumors, the outcomes with reirradiation using conventional RT are dismal, which poses a therapeutic challenge [7,8]. Recurrent ACC is not only challenging for its historically known radioresistance, but also for its frequently complex target volume (e.g. horseshoe shaped target volumes in case of perineural invasion), ⇑ Corresponding author at: Radiation Oncology Clinical Department, National Center for Oncological Hadrontherapy, Strada Campeggi 53, 27100 Pavia, Italy. E-mail address: [email protected] (B. Vischioni). https://doi.org/10.1016/j.radonc.2020.01.004 0167-8140/Ó 2020 Elsevier B.V. All rights reserved.

and the proximity to radiosensitive normal structures (e.g. tumors arising from minor salivary glands in paranasal sinuses). In the setting of reirradiation with photons, the therapeutic window is usually small (as normal structures in the proximity of target volume have already been irradiated) to deliver clinically meaningful tumoricidal doses to control the recurrent disease. In recent years high linear energy transfer (LET) RT with hadrons has been investigated to overcome some of the limitations of conventional RT in the treatment of salivary gland tumors [9]. In hadrontherapy or particle beam RT, high energy is released very precisely and selectively within the so called Bragg peak on the tumor target. Early investigations with hadrons, either with neutrons [10], carbon ions [11–13] or protons [12,13], have shown favorable outcomes after treatment of salivary gland tumors [14,15] High-LET RT has several radiobiological advantages over lowLET RT, in particular the higher radiobiological effectiveness due the increased energy deposition per unit length of track. Furthermore, compared to conventional RT, it has a low oxygen enhance-

173

B. Vischioni et al. / Radiotherapy and Oncology 145 (2020) 172–177

ment ratio (OER), less sensitivity of tumoricidal effect upon specific cell cycle phase, and cause a decreased ability of tumor cells to repair sublethal and potentially lethal damage [16]. Carbon ions beams have similar radiobiological advantages to neutrons, largely employed in the past for salivary gland cancer [10], but are characterized by superior dose distribution profile [14]. Unlike RT with neutrons, in carbon ion RT (CIRT) the high LET region can be conformed to the tumor [17], thus the therapeutic ratio might be potentially increased in CIRT. Clinical evidence for reirradiation with CIRT of recurrent salivary gland tumors is poor, and derived from the experience in the radical setting. The National Center for Oncological Hadrontherapy (CNAO) has started clinical activity with pts in 2011 with protons, and in 2012 with carbon ions active spot scanning [18]. In this study we report our first experience of retreatment of recurrent salivary gland cancers at CNAO with CIRT spot scanning. Materials and methods Patient population From February 2013 to September 2016, 51 consecutive pts with recurrent salivary gland tumors were retreated with CIRT at CNAO for relapse after primary treatment including at least one previous photon RT course. CIRT enrollment and treatment were performed according to the prospective phase II clinical trial for recurrent head and neck tumors Phase II clinical trial: carbon ion radiotherapy for head and neck relapsed tumors (trial n: CNAO S14/2012C, version: 1.0; 27.11.2012), approved by CNAO Ethical Review Board in 2012. The treatment protocol has been extended by the Italian National Institute of Health (ISS) until the European Community (EC) label certification marking of the CNAO particle accelerator for pt treatment in 2016 [19]. All enrolled pts gave written informed consent for CIRT, and permission for the use of their anonymized data for research purposes. Evaluation of pts prior to CIRT included history, physical examination, basic bloodwork, axial imaging with computed tomography (CT) or magnetic resonance imaging (MRI) of the head and neck, along with detailed review of prior RT dosimetry records. Tumor staging was performed on initial recurrence diagnostic scans prior to CIRT to the most recent TNM staging classification and clinical staging followed the American Joint Committee of Cancer and International Union Against Cancer (AJCC/UICC) guidelines (8° Edition) [20]. CIRT treatment and patient follow up For simulation CT scans a slice thickness of 0.2 cm was adopted. CNAO 3 Tesla MRI images (acquired with the same patient head rest and same mask of simulation CT) were registered to simulation CT in order to provide more information for target and organ at risks (OARs) delineation. The structures contoured for planning were the macroscopic tumor detected at imaging defined as the gross tumor volume (GTV), the GTV plus 0–5 mm margin, and, in a few cases, the microscopic extension of the tumor along the nerves as the clinical target volume (CTV), and a 2-mm CTV expansion to create the planning target volume (PTV). For each patient, Digital Imaging and Communications in Medicine (DICOM) files of previous RT or dose data of previous RT extracted from printed CT images with isodose curves were collected and analyzed, in order to produce cumulative dose volume histograms (DVH), and better estimate the risk of side effects of the new cumulative plan. An estimate of the cumulative biological equivalent dose (EQD2) from the prior and planned reirradiation was performed using a conservative a/b-ratio of 2 for all OARs as previously reported [21]. Constraints to OARs were set by using the QUANTEC reports, except for a few limiting–doses of CIRT for

optic nerve, brain stem and spinal cord, which were derived from previous experience from the National Institute of Radiological Sciences (NIRS) in Japan [22]. CIRT prescription doses varied based on examination of photon RT plan and site of recurrence, in a relative biological effectiveness (RBE)-weighted dose range between 45 and 68.8 Gy [RBE]. CIRT median dose was 60 Gy [RBE], delivered in 3.0–5.0 Gy [RBE]/fraction, 4 fractions per week, for a median BED (Biological Equivalent Dose) of 72 Gy (range 37.5–138.1). Fractionation schedule were decided for each patient based on pathology, tumor size, tumor location, or previous photon dose, in order to maximize expected tumor control and better spare OARs (Table 1). RBE was calculated with the local effect model version 1 [23] using the Syngo RT (Siemens Healthcare, Erlangen, Germany) Treatment

Table 1 Major patients and radiotherapy characteristics. Patients and treatment characteristics Sex Male Female Prior surgery None One Two Three Four Histology Adenoid cystic carcinoma Mucoepidermoid carcinoma Myoepithelial carcinoma Carcinoma ex pleomorphic adenoma Mucinous adenocarcinoma Ductal adenocarcinoma Site of retreatment Parotid Nasal cavity Nasopharynx Mandible Maxillary sinus Hard palate Ethmoid Para-pharyngeal space Oropharynx Lacrimal gland Soft palate Tongue Retromolar trigone Pterygopalatine fossa Reirradiation stage rcT2 rcT3 rcT4a rcT4b

N (%) 27 (53) 24 (47) 1 (2) 10 (19.6) 23 (45.1) 10 (19.6) 7 (11.7) 38 (74.5) 6 (11.8) 3 (5.8) 2 (3.9) 1 (2) 1 (2) 17 (33.3) 5 (9.8) 3 (5.9) 2 (3.9) 5 (9.8) 3 (5.9) 3 (5.9) 3 (5.9) 1 (2) 2 (3.9) 1 (2) 1 (2) 1 (2) 4 (7.8) 1 (2) 5 (9.8) 26 (51) 19 (37.2)

rcN0 rcN1 rcN2b

46 (90.2) 4 (7.8) 1 (2)

M0 M1 Prior RT courses One Two CIRT fractionation scheme 3.0 Gy [RBE]/fr  15 fr 3.0 Gy [RBE]/fr  16 fr 3.0 Gy [RBE]/fr  18 fr 3.0 Gy [RBE]/fr  19 fr 3.0 Gy [RBE]/fr  20 fr 3.0 Gy [RBE]/fr  22 fr 3.75 Gy [RBE]/fr  16 fr 4.0 Gy [RBE]/fr  14 fr 4.0 Gy [RBE]/fr  15 fr 4.0 Gy [RBE]/fr  16 fr 4.3 Gy [RBE]/fr  16 fr 5.0 Gy [RBE]/fr  12 fr

45 (88.2) 6 (11.8) 46 (90.1) 5 (9.9) 1 (2) 3 (5.8) 10 (19.6) 1 (2) 15 (29.4) 2 (3.9) 1 (2) 1 (2) 6 (11.7) 1 (2) 9 (17.6) 1 (2)

174

Reirradiation of salivary gland tumors

Planning System (TPS). All CIRT treatments were performed with spot scanning beam delivery method. Regular follow up took place at our Institution or the referring center, including MRI imaging of the head and neck and clinical visit, each 3 months post completion of treatment for the first 2 years, and then at 6-months intervals. Acute and late toxicities were scored according to the National Cancer Institute’s Common Terminology Criteria for Adverse Events CTCAE version 4. Tumor response assessment on follow up MRI scans was performed according to the RECIST criteria and clinical evaluation. Statistical analyses Chi-square and Kruskal–Wallis tests for nominal and ordinal variables were used to compare groups of pts ranked according to their clinical characteristics. Time to event data was calculated from CIRT start to last follow up or death. Kaplan–Meier curves were compared using the log-rank tests. Univariate and multivariate analysis were performed to evaluate impact of different variables on pts treatment outcomes and survival. Hazard ratios (HR) with 95% confidence limits (CI) were calculated using the Cox proportional hazard model. Local control (LC) was defined as the absence of further tumor growth following CIRT or the absence of further tumor growth following best response of the treated lesion(s). Progression-free survival (PFS) was defined as the absence of locoregional or distant failure or death of any cause. Overall survival (OS) was defined till death or the last follow up. All tests were two-tailed, and the level of significance was set at 0.05. Statistical analyses was performed with IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp and SAS 9.4. Results Pts median age was 60 years (range: 30–85 years). In Table 1 are summarized major pts and RT characteristics. In Fig. 1 a representative case is shown from our series treated with CIRT. All pts had gross disease at the time of CIRT, in 42 pts (82.4%) detected in the high dose area of prior photon based RT, and in 9 pts

(17.6%) at the field edge of the target volumes of the first RT plan and not receiving 95% of the prescribed dose to 95% of the recurrent volume. Median treated GTV was 28.58 cc (range: 1.75 cc–205.54 cc), median CTV 76.56 cc (range 2.74 cc–865 cc). Median dose of prior photon based RT was 60 Gy (range: 24–78 Gy). Median time interval between prior RT and CIRT was 6.33 years (range: 1.08–20 years). Median follow up time was 19 months (range: 2–57 months). At the 6 months follow up visit, 4 pts had a complete response (7.8%), 15 a partial remission (29.4%) and 25 stable disease (49.0%). At the time of last follow up, 21 pts (41.2%) had the local tumor controlled, and 30 pts (58.8%) had tumor progression. Of these 30 pts with disease progression, 12 (40%) had both local and distant metastasis, 17 (56.7%) only local relapse and 1 (3.3%) local and nodal progression. All pts were actively followed up after the progression. All but 2 received best supportive care, one patient received another course of CIRT and another one surgical removal of the relapsed lesion. Overall, 17 pts (33%) developed distant metastases at the time of last follow up. Only 5 cases had metastasis without local progression (29.4%) whereas in 12 pts (70.6%) local disease progression was diagnosed together with distant failure (p = 0.09). Tumor progression was located within the carbon ion fields in 18 of 30 pts (60%), 4 pts (13%) had disease progression outside the carbon ion fields, and 8 pts (27%) had progression at field border, at site towards critical OARs. At the time of last follow up, 31 pts (60.8%) were alive and 20 pts (39.2%) dead. Estimated PFS (actuarial) at one and two years were 71.7% and 52.2%, respectively. Estimated OS (actuarial) at one and two years were 90.2% and 64.0%, respectively (Fig. 2). The median PFS was 25.00 (95% CI: 14–34) months. Median OS was not reached as more than 50% of pts were alive at the last follow up. With the Kaplan–Meier analyses, among other pts characteristics females had better PFS (p = 0.07) and OS (p = 0.005). Using receiver operating curve analyses with the Youden’s index, we calculated a cut-off value of 62 cc of GTV above which the median survival was significantly lower after treatments. The median PFS in pts with GTV less than 62 cc and more than 62 cc were 32 (95% CI: 18–55) and 9 months (95%CI: 2–17), respectively (log rank test,

Fig. 1. Images of a 75-years-old female patient with a nasopharyngeal ACC relapsed at the pterygopalatine fossa 7 years after first postoperative RT course of 60 Gy in 30 fractions. A, B, CIRT plan of 60 Gy [RBE], delivered in 20 fractions of 3.0 Gy [RBE]/fraction, 4 fractions per week with IMPT (Intensity Modulated Particle Therapy) technique depicted on axial (A) and coronal (B) planes of the simulation CT scan. In red is shown contour for GTV volume, in green for CTV and in violet for PTV. C, D, T1-weighted postcontrast MRI images of the relapsed tumor before CIRT, on axial (C) and coronal (D) planes. E, F, T1-weighted post-contrast MRI images of the relapsed tumor 4 months after CIRT, on axial (E) and coronal (F) planes, showing partial response to treatment.

175

B. Vischioni et al. / Radiotherapy and Oncology 145 (2020) 172–177

In multivariate Cox proportional hazard model, the significance of the GTV for OS, PFS and LC was maintained even when adjusting for other covariates such as age (Table 3). Gender was also found to be significant in multivariate analysis for OS, same as reirradiation interval with decreasing HR for longer intervals between CIRT reirradiation and previous RT. A trend for significance of the delivered CIRT dose for LC could be assessed (p = 0.09). Table 4 reports scoring of pts acute and late toxicity at last follow up. Most common acute toxicities were erythema (18 pts, 35.3%) and mucositis (10 pts, 19.6%). During follow up, 5 pts (9.8%) developed visual deficit (1 G1, 1 G2 and 3 G3), 2 pts (3.9%) developed xerostomia G1, 9 pts (17.6%) neuropathy (3 G1, 5 G2 and 1 G3), and 8 pts (15.7%) developed trismus (1 G1, 3 G2 and 4 G3). Furthermore, 1 patient developed mandibular necrosis G1, and another one developed G1 brain necrosis.

Discussion

Fig. 2. Local control (LC) and overall survival (OS) following reirradiation with CIRT in a series of inoperable recurrent salivary gland tumors treated at CNAO.

p = 0.0001). A similar pattern was seen in OS where the median OS was not reached in those pts with GTV less than 62 cc, and was 16 months (95%CI: 6–27) in those with GTV bigger than 62 cc (log rank test, p = 0.0001). At univariate analysis pts with tumor volume less than 62 cc, longer reirradiation interval from previous RT (pts ranked according to median time interval between CIRT and previous RT course), and female gender, were associated with superior OS (significant for tumor volume and gender, with a trend for significance for time interval; Table 2). Pts with GTV larger than 62 cc were 5.371 times likelier to die over the next time interval than a patient with a small GTV (<62 cc) (95% CI: 2.02–14.28). Similar HR were found for PFS and LC.

Here we report outcome and toxicity after CIRT at CNAO on a series of recurrent salivary gland pts inoperable and heavily pretreated with previous RT courses and/or surgeries. From the beginning of the clinical activity up to the end of 2016, at CNAO all pts were treated under prospective protocols [19]. For reirradiation setting, 2 competitive protocols were ongoing, one with proton RT and the other with CIRT. Proton RT was used as a first option, CIRT was used for bad prognosis and radioresistant histology tumors with a poor response to low LET RT (sarcoma, melanoma and salivary gland tumors) [21]. Although the adverse features of the tumor cases treated with CIRT at CNAO, overall, our data are in line with Jensen et al. [24] on recurrent adenoid cystic carcinomas treated with CIRT, 83% of which inoperable. Furthermore, though distant failure was common, achieving good LC with CIRT at CNAO resulted in a trend towards significantly lesser distant metastasis, suggesting that LC was important in the recurrent setting to prevent development of systemic recurrence. At present, besides particle therapy, different approaches are routinely considered for the management of recurrent salivary gland tumors, such as salvage surgery (for operable tumors) with or without intraoperative radiotherapy (IORT) [7], and Stereotactic Body Radiation Therapy (SBRT) [25]. Chen et al. [7] have reported experience of recurrent operable salivary gland tumors treated

Table 2 Univariate analysis for prognostic value of major patients and treatment characteristics. Model covariates Gross tumor volume (<62 cc vs  62 cc) Nodal disease (N0 vs N+) Re-radiation interval

Sex (Males vs females) M1 disease before CIRT

Age (<60 years vs 60 years) CIRT radiation dose

HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value

OS

PFS

LC

5.371 2.019–14.284 0.0008 0.460 0.062–3.441 0.45 0.886 0.768–1.021 0.09 4.222 1.410–12.641 0.01 0.317 0.042–2.379 0.26 1.394 0.577–4.369 0.46 0.965 0.894–1.041 0.36

4.575 2.021–10.358 0.0003 0.807 0.245–2.663 0.72 0.947 0.855–1.048 0.29 2.282 1.104–4.718 0.03 0.798 0.278–2.292 0.68 1.511 0.756–3.021 0.24 0.961 0.906–1.018 0.18

4.593 2.931–10.920 0.0006 0.878 0.264–2.919 0.83 0.973 0.880–1.076 0.59 1.971 0.930–4.175 0.08 0.876 0.303–2.537 0.81 1.662 0.798–3.460 0.17 0.946 0.889–1.006 0.08

176

Reirradiation of salivary gland tumors

Table 3 Multivariate analyses for prognostic value of major patients and treatment characteristics. Model covariates Gross tumor volume (<62 cc vs 62 cc) Nodal disease (N0 vs N+) Re-radiation interval

Sex

M1 disease before CIRT

Age (<60 years vs 62 years) CIRT radiation dose

HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value HR (95% CI) P-value

OS

PFS

LC

4.831 1.645–14.187 0.004 0.288 0.031–2.690 0.27 0.836 0.709–0.987 0.03 4.325 1.250–14.959 0.02 0.156 0.018–1.361 0.09 1.387 0.441–4.363 0.58 0.986 0.909–1.069 0.73

5.597 2.027–15.455 0.0009 0.445 0.110–1.793 0.25 0.954 0.831–1.095 0.50 1.808 0.799–4.091 0.16 0.740 0.245–2.238 0.59 1.146 0.499–2.633 0.75 0.956 0.898–1.017 0.16

6.683 2.144–20.834 0.001 0.390 0.088–1.741 0.22 0.998 0.863–1.153 0.98 1.457 0.622–3.412 0.39 0.863 0.282–2.640 0.80 1.196 0.511–2.799 0.68 0.943 0.882–1.008 0.09

Table 4 Acute and late toxicity at last follow up.

G0 G1 G2 G3

ACUTE TOXICITY N (%)

LATE TOXICITY N (%)

11 (21.5) 19 (37.3) 19 (37.3) 2 (3.9)

14 (27.5) 9 (18) 19 (37) 9 (17.5)

with salvage surgery with or without IORT. Previous RT had been delivered in 82% of pts. At the time of salvage, 37% of pts received IORT to a median dose of 15 Gy. The 1-, 3-, and 5-year estimates of LC after salvage surgery were 88%, 75%, and 69%, respectively. Distant metastasis was the most common pattern of subsequent failure, occurring in 42% of pts. The 5-year OS was 34%. Despite pts with inoperable multiple recurrent tumors (80% of pts recurred after more than two prior surgeries), oncologic outcomes after CIRT at CNAO are comparable to operable recurrent tumors. Karam et al. [25] have reported experience of recurrent salivary gland tumors in 18 pts treated with SBRT. Most pts did not undergo surgical resection, and among those that did, all had positive margins. The median SBRT dose was 30 Gy given in five fractions, with a median cumulative dose of 91.1 Gy. The 2-year OS, PFS, and LC rates were 39%, 24%, and 53%, respectively. CIRT at CNAO, in a similar patient population, resulted in superior oncologic outcome with estimated 2-year OS and PFS rates of 64% and 52.2%, respectively. Yamazaki et al. [26] has recently reported, in a multicentric retrospective study in Japan, better outcome with particle therapy compared to photon RT when reirradiating recurrent head and neck tumors. Prognostic factors for better OS were primary site (nasopharynx), GTV < 40 cc and prescribed doses (EQD2) > 40 Gy. In our study, only GTV volume has been shown to significantly influence LC, PFS and OS at both univariate and multivariate analysis, whereas only a trend for significance was assessed for prognostic value of delivered CIRT dose and LC. At multivariate analysis other prognostic factors for OS in our series were time interval from previous RT and sex, with female gender and longer time interval from previous RT associated with longer survival. Concerning sex, the peculiar mixture of cases and the small number of pts in our series does not allow to draw general conclusions on its impact on prognosis since not reported elsewhere in reirradiation setting with CIRT. On the other side,

reirradiation time interval has already been shown to affect patient outcome and prognosis in Hayashi et al. [22], where an interval 24 predicted longer PFS and OS at multivariate analysis in a series of 46 pts with recurrent head and neck malignancies treated with CIRT after an initial course of CIRT. In this regard, the late onset of the relapse from the first RT course might underlie the slow growing tendency of the primary tumor, and thus the good prognosis. Toxicity was not a major issue for the pts treated with CIRT at CNAO. Only 3.9% of the pts had acute G3 toxicity (mucositis and erythema), none had higher than G3 toxicity nor any pts had to interrupt treatment course because of toxicity. Late toxicity in our series has been very moderate as well. Reported G3 toxicities were visual deficit (as expected by the cumulative doses to the optic structures), neuropathy and trismus. None of our pts developed toxicity grade G4 and G5, differently to other published series treated with CIRT [22,24,26,27]. In these papers with similar follow up time, major pts hazards are soft tissue necrosis and carotid blowout similar to series treated with cyberknife photon SBRT [26,27]. At CNAO we might have been able to avoid the most dangerous treatment sequelae by applying specific dose constraints to the carotid artery with the carotid sparing technique [21], and personalizing treatment total dose and fractionation based on histology and imaging pts feature. On the other side, retreated tumors in our series were mainly located at the parotid gland, differently from other published series with recurrent tumors located mainly at the nasopharynx or nasal cavity [22,24]. In this regard, Yamazaki et al. reported the nasopharyngeal location for reirradiation as significant predictor of G3 or worse toxicity at univariate analysis, PTV volume >40 cm3 affected toxicity at both univariate and multivariate analysis [26]. Impact of target volume and reirradiation location on toxicity is not surprising since dose escalation in critical areas close to the base of the skull might require increased caution, and then also imply higher toxicity risk if aiming at the goal of controlling the tumor, as highlighted in Hayashi et al. where the authors reported a late toxicity rate of G3 and higher of 37.5% [22]. In our series, 87% of relapses were infield, or at the field border towards critical structures of CIRT treatment plan. Concerning dose constraints to critical OARs irradiated with CIRT, it is well know that it is not possible to extend the experience with photon RT adjusted for hypofractionation, or of CIRT facilities abroad, to CIRT delivered at CNAO, since the RBE dependence on tissue specificity, fractionation, radiation quality, besides issues linked to the dosimetric model used in treatment planning [17,28]. Considering

B. Vischioni et al. / Radiotherapy and Oncology 145 (2020) 172–177

the low rate of toxic events discussed previously and the not negligible infield relapse rate, we might infer that our constraints towards normal critical structures might have been too conservative, with the risk to underdose the tumor volume, besides underlining the well-known intrinsic radioresistance of salivary gland tumors requiring steep dose escalation to achieve adequate tumor control. For the future, in order to increase CIRT treatment outcome and in consideration of the good tolerability of CIRT in the reirradiation setting at CNAO, we are planning to increase CIRT prescribed total dose also in the recurrent cases after a previous photon RT course. In fact, in the radical setting, CIRT dose prescribed at CNAO for salivary gland tumors is currently 65.6–68.8 Gy [RBE], 4.1–4.3 Gy [RBE]/fraction, 4 fractions per week [19], for a BED dose of 155.2–167.4 Gy [RBE], well below the median BED dose of the present series of pts retreated with CIRT. In conclusion, in the reirradiation setting, CIRT is effective in controlling local progression of recurrent salivary gland tumors along with acceptable rates of acute and late toxicity when surgery is not an option. CIRT appears to be a reasonable alternative for palliative therapy for pts with inoperable, multiply recurrent, previously irradiated salivary gland cancer with good performance status. Conflict of interest None. Acknowledgement Drs Hasegawa was supported by the research grant from Associazione Italiana per la Ricerca sul Cancro (AIRC): IG-14300. References [1] Shen C, Xu T, Huang C, Hu C, He S. Treatment outcomes and prognostic features in adenoid cystic carcinoma originated from the head and neck. Oral Oncol 2012;48:445–9. [2] Chen AM, Bucci MK, Weinberg V, Garcia J, Quivey JM, Schechter NR, et al. Adenoid cystic carcinoma of the head and neck treated by surgery with or without postoperative radiation therapy: prognostic features of recurrence. Int J Radiat Oncol Biol Phys 2006;66:152–9. [3] Chen AM, Granchi PJ, Garcia J, Bucci MK, Fu KK, Eisele DW. Local-regional recurrence after surgery without postoperative irradiation for carcinomas of the major salivary glands: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 2007;67:982–7. [4] Mendenhall WM, Morris CG, Amdur RJ, Werning JW, Villaret DB. Radiotherapy alone or combined with surgery for salivary gland carcinoma. Cancer 2005;103:2544–50. [5] Le QT, Birdwell S, Terris DJ, Gabalski EC, Varghese A, Jr Fee We, et al. Postoperative irradiation of minor salivary gland malignancies of the head and neck. Radiother Oncol 1999;52:165–71. [6] Garden AS, el-Naggar AK, Morrison WH, Callender DL, Ang KK, Peters LJ. Postoperative radiotherapy for malignant tumors of the parotid gland. Int J Radiat Oncol Biol Phys 1997;37:79–85.

177

[7] Chen AM, Garcia J, Bucci MK, Chan AS, Kaplan MJ, Singer MI, et al. Recurrent salivary gland carcinomas treated by surgery with or without intraoperative radiation therapy. Head Neck 2008;30:2–9. [8] Pederson AW, Haraf DJ, Blair EA, Stenson KM, Witt ME, Vokes EE, et al. Chemoreirradiation for recurrent salivary gland malignancies. Radiother Oncol 2010;95:308–11. [9] Orlandi E, Iacovelli NA, Bonora M, Cavallo A, Fossati P. Salivary gland. Photon beam and particle radiotherapy: present and future. Oral Oncol 2016;60:146–56. [10] Douglas JG, Koh W-J, Austin-Seymour M, Laramore GE. Treatment of salivary gland neoplasms with fast neutron radiotherapy. Arch Otolaryngol Head Neck Surg 2003;129:944–8. [11] Mizoe J-E, Tsujii H, Kamada T, Matsuoka Y, Tsuji H, Osaka Y, et al. Dose escalation study of carbon ion radiotherapy for locally advanced head-andneck cancer. Int J Radiat Oncol Biol Phys 2004;60:358–64. [12] Morimoto K, Demizu Y, Hashimoto N, Mima M, Terashima K, Fujii O, et al. Particle radiotherapy using protons or carbon ions for unresectable locally advanced head and neck cancers with skull base invasion. Jpn J Clin Oncol 2014;44:428–34. [13] Jäkel O, Smith AR, Orton CG. The more important heavy charged particle radiotherapy of the future is more likely to be with heavy ions rather than protons. Med Phys 2013;40:090601. [14] Stannard C, Vernimmen F, Carrara H, Jones D, Fredericks S, Hille J, et al. Malignant salivary gland tumours: can fast neutron therapy results point the way to carbon ion therapy?. Radiother Oncol 2013;109:262–8. [15] Laramore GE, Krall JM, Griffin TW, Duncan W, Richter MP, Saroja KR, et al. Neutron versus photon irradiation for unresectable salivary gland tumors: final report of an RTOG-MRC randomized clinical trial. Int J Radiat Oncol Biol Phys 1993;27:235–40. [16] Loeffler JS, Durante M. Charged particle therapy–optimization, challenges and future directions. Nat Rev Clin Oncol 2013;10:411–24. [17] Schulz-Ertner D, Jäkel O, Schlegel W. Radiation therapy with charged particles. Semin Radiat Oncol 2006;16:249–59. [18] Rossi S. The National Centre for Oncological Hadrontherapy (CNAO): status and perspectives. Phys Med 2015;31:333–51. [19] Fossati P, Matsufuji N, Kamada T, Karger CP. Radiobiological issues in prospective carbon ion therapy trials. Med Phys 2018;45:e1096–110. [20] Brierley JD, Gospodarowicz MK, Wittekind C, editors. The TNM classification of malignant tumours. Oxford: Wiley Blackwell; 2017. [21] Dale JE, Molinelli S, Ciurlia E, Ciocca M, Bonora M, Vitolo V, et al. Risk of carotid blowout after reirradiation with particle therapy. Adv Radiat Oncol 2017;2:465–74. [22] Hayashi K, Koto M, Ikawa H, Hagiwara Y, Tsuji H, Ogawa K, et al. Feasibility of Re-irradiation using carbon ions for recurrent head and neck malignancies after carbon-ion radiotherapy. Radiother Oncol 2019;136:148–53. [23] Kramer M, Scholz M. Treatment planning for heavy-ion radiotherapy: calculation and optimization of biologically effective dose. Phys Med Biol 2000;45:3319–30. [24] Jensen AD, Poulakis M, Nikoghosyan AV, Chaudhri N, Uhl M, Münter MW, et al. Re-irradiation of adenoid cystic carcinoma: analysis and evaluation of outcome in 52 consecutive patients treated with raster-scanned carbon ion therapy. Radiother Oncol 2015;114:182–8. [25] Karam SD, Snider JW, Wang H, Wooster M, Lominska C, Deeken J, et al. Reirradiation of recurrent salivary gland malignancies with fractionated stereotactic body radiation therapy. J Radiat Oncol 2012;1:147–53. [26] Yamazaki H, Demizu Y, Okimoto T, Ogita M, Himei K, Nakamura S, et al. Reirradiation for recurrent head and neck cancers using charged particle or photon radiotherapy. Strahlenther Onkol 2017;193:525–33. [27] Yamazaki H, Demizu Y, Okimoto T, Ogita M, Himei K, Nakamura S, et al. Comparison of re-irradiation outcomes for charged particle radiotherapy and robotic stereotactic radiotherapy using cyberknife for recurrent head and neck cancers: a multi-institutional matched-cohort analysis. Anticancer Res 2016;36:5507–14. [28] Dale JE, Molinelli S, Vitolo V, Vischioni B, Bonora M, Magro G, et al. Optic nerve constraints for carbon ion RT at CNAO – reporting and relating outcome to European and Japanese RBE. Radiother Oncol 2019;140:175–81.