- Email: [email protected]
Toxicities Following Stereotactic Ablative Radiotherapy Treatment of Locally-Recurrent and Previously Irradiated Head and Neck Squamous Cell Carcinoma
Toxicities Following Stereotactic Ablative Radiotherapy Treatment of Locally-Recurrent and Previously Irradiated Head and Neck Squamous Cell Carcinoma Kimmen Quan, MD, FRCPC, Karen M. Xu, BS, Yongqian Zhang, PhD, David A. Clump, MD, PhD, John C. Flickinger, MD, Ron Lalonde, PhD, Steven A. Burton, MD, and Dwight E. Heron, MD, MBA, FACRO, FACR Stereotactic ablative radiotherapy (SABR) with concomitant cetuximab is an effective treatment option for previously irradiated, locally recurrent squamous cell carcinoma of the head and neck. Its local control and overall survival are similar to those of other available treatment options. Each retreatment depends heavily on the prior treatment and every patient is a special case. Based on the experience of our institution and previously published studies, for patients who receive concomitant cetuximab with a median prior radiation therapy dose of 70 Gy, we recommend a total dose of 40-44 Gy delivered in 5 fractions on alternating days over 1-2 weeks. However, Grade 2 or 3 toxicities are not uncommon. Therefore, in this review, we also report a pilot study that applies a normal tissue complication probability dose-response model to estimate the probability of toxicities in locally recurrent squamous cell carcinoma of the head and neck reirradiated with SABR. Although this dose-response model includes concurrent targeted therapy and no comparable model yet exists for SABR without it, complication rates without concurrent biological therapy or chemotherapy should be no higher than those described here. Semin Radiat Oncol 26:112-119 C 2016 Elsevier Inc. All rights reserved.
T
he locoregional recurrence rate in squamous cell carcinoma of the head and neck (SCCHN) following definitive and adjuvant chemoradiation therapy remains substantial at approximately 20%-30%,1,2 despite major improvements in the multimodality treatment of SCCHN. Radiation Therapy Oncology Group protocols reported locoregional recurrences as first failure sites in 50%-60% of both human papillomavirus-positive and human papillomavirus-negative subsets.3 An additional problem is the development of second primary head and neck cancers. The concept of field cancerization was first coined by Slaughter in 1953, who noticed the presence of histologically abnormal tissue surrounding oral
Department of Radiation Oncology, University of Pittsburgh Cancer Institute, Pittsburgh, PA. Conflicts of interest: none. Address reprint requests to Dwight E. Heron, MD, FACRO, Department of Radiation Oncology, University of Pittsburgh Cancer Institute, 5230 Centre Ave, Pittsburgh, PA 15232. E-mail: [email protected]
112
http://dx.doi.org/10.1016/j.semradonc.2015.11.007 1053-4296//& 2016 Elsevier Inc. All rights reserved.
squamous cell carcinoma. He proposed that the entire mucosa undergoes histological changes when exposed to environmental carcinogens such as tobacco and alcohol and is therefore susceptible to developing multiple primary malignancies.4 Based on this concept, the risk of a second primary malignancy for successfully treated SCCHN associated with tobacco and alcohol use is at least 1% per year.5 As locoregional progression is the most common cause of death,6 achieving local control in previously irradiated, locally recurrent disease may have a potentially significant effect on survival. Good local control also improves quality of life as uncontrolled tumor growth is associated with significant pain, bleeding, and impairment of essential functions such as swallowing, speaking, and breathing. For patients with resectable SCCHN recurrence, surgical salvage is the standard treatment.6 Unfortunately, many patients with previously irradiated locally recurrent SCCHN (rSCCHN) have tumors that are unresectable and reirradiation with or without chemotherapy becomes the only potentially
Toxicities following stereotactic ablative radiotherapy treatment curative salvage option. However, prospective randomized clinical trials studying conventional reirradiation techniques with chemotherapy reported significant toxicities with Grade 3 or higher acute toxicity rates of 64%-78% and Grade 3 or higher late toxicity rates of 22%-38%.7,8 Even modern intensity-modulated radiation therapy (IMRT) is associated with high rates of mucosal and soft tissue toxicities.9,10 Stereotactic ablative radiotherapy (SABR), owing to its great conformity and precision in dose delivery, has been associated with fewer normal tissue toxicities than conventional radiotherapy and IMRT. Furthermore, SABR requires much shorter treatment time (around 1-2 weeks) compared with IMRT, which can take 6-7 weeks. Since 2003, patients with previously irradiated rSCCHN have been treated on protocols with SABR ⫾ cetuximab at our institution.11-13 Although the treatment was in general well tolerated, we did observe some Grade 2-3 toxicities and occasional Grade 4 toxicities. The purpose of this review is to evaluate the toxicities of SABR in treating locally rSCCHN with previous irradiation and report results of a pilot study that applies the normal tissue complication probability (NTCP) dose-response model to study the dose-volume relationship to toxicities in previously irradiated, locally rSCCHN treated with SABR.
Toxicities of SABR in Reirradiating Locally rSCCHN Prospective Clinical Trials Several randomized, prospective clinical trials have investigated the use of SABR in reirradiating locally rSCCHN, as summarized in Table 1. Heron et al14 reported the first Phase I dose-escalation trial of reirradiating with SABR. Overall, 25 patients with locally rSCCHN, median tumor volume ¼ 44.8 (4.2-217) cm3, were retreated using SABR in 5 fractions with a total dose of 25-44 Gy, in 5 dose groups. The mean prior dose in each escalated dose group ranged from 66.0-69.6 Gy in 3036 fractions.14 Nobody received concurrent cetuximab or chemotherapy. The median time to disease progression after completion of SABR was 4 months and the median overall survival (OS) after completion of SABR was 6 months. No Grade 3 or 4 or dose-limiting toxicities occurred. In all, 2 patients experienced Grade 1 mucositis, 1 patient with Grade 2 dysphagia and 1 patient with Grade 1 hyperpigmentation. Selfreported quality of life was not significantly affected by SABR treatment. In 2012, Comet et al15 conducted a prospective study that included 40 patients with inoperable or new primary head and neck cancer in a previously irradiated area. Overall, 50% of these tumors were squamous cell carcinoma. The median prior RT dose was 66 Gy and median tumor diameter was 29 mm. Overall, 70% had previous surgery and 57% had received chemotherapy during their initial treatment. SABR was delivered in 6 fractions to the 85% isodose line with a total dose of 36 Gy. In this study, 15 patients received concurrent cetuximab and 1 received concurrent cisplatin. The median progression-free survival was 8.8 months (95% CI: 5.2-11.7
113 months) and the median OS was 13.6 months. The median follow-up was 25.6 months. No Grade 4 toxicity was observed. Of all, 4 (10.3%) patients developed Grade 3 toxicities including mucositis, dysphagia, induration and fibrosis, 3 of whom received concurrent cetuximab. In all, 8 (20.6%) patients experienced Grade 2 toxicities and 10 (25.6%) experienced Grade 1 toxicities. Iwata et al16 reported a prospective protocol-based study that included 51 patients with previously irradiated recurrent nasal or paranasal carcinoma. Most patients were reirradiated with a total dose of 30 Gy in 3 fractions or 35 Gy in 5 fractions. The median prior RT dose was 60 Gy (range: 40-70 Gy). The median interval between initial radiation and reirradiation was 18 months (range: 2.5-132 months). The median time-to-local progression after completion of SABR was 9.5 months and the median OS after completion of SABR was 14.5 months. Grade 3 or higher adverse events were observed in 23% of patients. A total of 2 patients developed Grade 4 dermatitis and 1 patient developed Grade 4 soft tissue necrosis. Of all, 2 brain necrosis patients presented with clinical symptoms. Lartigau et al17 reported a multiinstitution Phase II study that included 56 patients with previously irradiated, locally rSCCHN. SABR was delivered in 6 fractions to the 85% isodose line with a total dose of 36 Gy. All patients received concomitant cetuximab. The median prior RT dose was not reported. The median progression-free survival after completion of SABR was 7.1 months (95% CI: 5.5-8.9 months) and the median OS after completion of SABR was 11.8 months with a 1-year OS rate of 47.5% (95% CI: 30.8%-62.4%). The median follow-up was 11.4 months. A total of 18 patients developed Grade 3 toxicities including mucositis, dysphagia, induration, and fibrosis. More recently, Vargo et al18 reported a prospective Phase II trial of 50 patients with inoperable locally rSCCHN with a previous RT dose Z60 Gy and tumor volumes of 3.6-209.2 (median ¼ 36.5) cm3. The median time from initial radiation to reirradiation was 18 months (range: 3-423 months). The median prior RT dose was 70 Gy (range: 52.5-118.2 Gy). All patients received concomitant cetuximab. SABR delivered a total dose of 40-44 Gy in 5 fractions on alternating days over 12 weeks. The median OS was 10 months (95% CI: 7-16 months) with a 1-year local progression-free survival rate of 60% (95% CI: 44%-75%). The median follow-up was 18 months (range: 10-70 months). Of all, 6% of patients developed Grade 3 acute toxicities including 1 mucositis, 1 dysphagia and 1 skin rash. A total of 6% of patients developed Grade 3 late toxicities including 1 dysphagia (45 cm3 tonsil tumor), and 2 aerodigestive fistulae (45 cm3 stoma/submental and 135 cm3 base of tongue). There was no Grade 4 or greater toxicity. In summary, SABR is a feasible option to re-irradiate locally rSCCHN. It is associated with a reasonable local control rate, OS and toxicity profile. However, Grade 2 or 3 toxicities are not uncommon after treatment. Most commonly seen toxicities include mucositis, dysphagia, and dermatitis.
K. Quan et al
18 50 (100%) 5 70 (52.5-118.2) 18 (3-423) 50 Vargo et al18
40-44
11.4 56 (100%) 6 36 38 Not reported Lartigau et al17 56
21 0 1-5 35 (20-41.5) 60 (40-70)
18 (2.5-132) 51 Iwata et al16
66
31.6 (7.9-263.4)
36 40 Comet et al15
6
15 (37.5%)
25.6
2 (%) Grade 1 mucositis, 1 (4%) Grade 2 dysphagia, and 1 (4%) Grade 1 hyperpigmentation No Grade 4 toxicity, 4 (10.3%) Grade 3 toxicities, 8 (20.6%) Grade 2 toxicities, and 10 (25.6%) Grade 1 toxicities Grade 3 or higher adverse events in 23% of patients, 2 (3.9%) Grade 4 dermatitis, and 1 (1.96%) Grade 4 soft tissue necrosis 18 (32.1%) Grade 3 toxicities: mucositis, dysphagia, induration and fibrosis 3 (6%) Grade 3 acute toxicities: 1 mucositis, 1 dysphagia and 1 skin rash. 3 Grade 3 (6%) late toxicities: 1 dysphagia, and 2 aerodigestive fistulas. No Grade 4 or greater toxicity Not Reported 0 5 64.7
Not Reported
40 (25-44) 25 Heron et al14
SABR Median Fraction (Range) SABR Median Time No. of Prior RT Patients Median Dose From RT to SABR, Median Dose (Gy) (Gy) (Range) Months (Range) (Range) Study
Table 1 Prospective Clinical Trials That Investigated Reirradiating Locally rSCCHN With SABR
Toxicities Median No. of Patients With Concurrent Follow-up (Months) Cetuximab (%)
114
Retrospective Studies Multiple retrospective studies investigated toxicities of SABR in treating locally rSCCHN with previous irradiation, as summarized in Table 2. Roh et al19 reported a study of 36 patients reirradiated with SABR for locally rSCCHN. A total dose of 1840 Gy was delivered in 3-5 fractions to the 65%-85% isodose line. The median prior RT dose was 70.2 Gy (range: 39.6134.4 Gy). The median follow-up was 17.3 months. Grade 3 acute toxicities were observed in 13 (36.1%) patients and late complications (1 bone necrosis and 2 soft tissue necrosis) were observed in 3 (8.3%) patients. Unger et al20 reported a study of 65 patients including 38 treated definitively and 27 treated for metastatic diseases. A total of 11 (17%) patients received concurrent cetuximab. The median initial RT dose was 66.6 Gy (range: 32.4-120.2 Gy). SABR delivered a median dose of 30 Gy (range: 21-35 Gy) in 2-5 fractions. The median OS for all patients was 12 months. No patients experienced Grade 4 acute toxicity. A total of 19 patients experienced Grade 1-3 acute toxicities, including mucositis, dermatitis, and nausea. A total of 6 (9%) patients experienced Grade 4 toxicities, including arterial bleeding, soft tissue necrosis, fistula formation and dysphagia. A single patient died from treatment-related complications. Cengiz et al21 reported 46 locally recurrent head-and-neck tumors irradiated with SABR. The most common histopathology was epidermoid carcinoma. The median prior RT dose was 61 Gy (range: 30-70 Gy). SABR delivered a median dose of 30 Gy (range: 18-35 Gy) in a median of 5 (range: 1-5) fractions. The median OS after completion of SABR was 11.9 months (range: 11.4-17.4 months). A single (2.2%) patient developed Grade 2 acute mucositis. Of all, 2 (4.4%) patients experienced Grade 3 acute dermatitis and mucositis. Overall, 6 (13.3%) patients developed long-term complications. A single patient had soft tissue necrosis and 1 patient developed mandibular necrosis. Of all, 2 (4.4%) patients developed Grade 3 dysphagia. Of all, 2 (4.4%) patients developed Grade 2 dysphagia. Vargo et al22 retrospectively reviewed 132 patients with previously irradiated, locally rSCCHN with median tumor volume of 30.9 (4.4-192.4) cm3. SABR was delivered in 5 fractions with a total dose of 44-50 Gy. A total of 55% of patients received concomitant cetuximab. The median prior RT dose was 68 Gy (range: 20.4-140 Gy) at a reirradiation interval of 23 months (range: 2-423 months). A total of 3 patients developed Grade 3 acute mucositis and 1 developed Grade 4 acute mucositis. A single patient developed Grade 3 acute dysphagia and 2 patients developed Grade 4 acute dysphagia. Of all, 3 patients developed Grade 3 skin rash and 1 developed Grade 4 skin rash. At a median follow-up of 6 months (range: 0-55 months), treatment duration o14 days was associated with significantly improved recurrence-free survival (37% vs 8% 1-year, p ¼ 0.041), in univariate but not multivariate analysis at the expense of increased late toxicity (8% vs 0%, p ¼ 0.017). Tumor volume 425 ml was associated with significantly more acute toxicity (75% vs 66%, p ¼ 0.017) in univariate and multivariate analysis but had no effect on late toxicity. 23 More recently, Kress et al retrospectively reviewed 85 patients with previously irradiated SCCHN treated with SABR.
17.3 5 (3-5) Kress 85 et al23
68 (32.4-120.2)
32 (1-324)
30 (16-41)
22 (25.9%)
6 5 23 (2-423) 68 (20.4-140) 132
44 (35-50)
72 (55%)
7 0 5 (1-5) 30 (18-35)
Roh et al19 Unger et al20 Cengiz et al21 Vargo et al22
38 (3.8-306) 61 (30-70) 46
2-5 66.6 (32.4-120.2) 26 (1.6-318) 65
30 (21-35)
11 (17%)
12
13 (36%) Grade 3 acute toxicities and 3 (8.3%) late complications: 1 bone necrosis and 2 soft tissue necrosis 19 (29%) Grade 1-3 acute toxicities. 6 (9%) Grade 4 toxicities. 1 died from treatment-related complications 1 (2.2%) Grade 2 acute mucositis. 2 (4.4%) Grade 3 acute dermatitis and mucositis. 6 (13.3%) late complications 3 (2.3%) Grade 3 acute mucositis and 1 (0.75%) Grade 4 acute mucositis. 1 (0.75%) Grade 3 acute dysphagia and 2 (1.5%) Grade 4 acute dysphagia. 3 (2.3%) Grade 3 skin rash and 1 (0.75%) Grade 4 skin rash 2 (2.4%) Grade 3 acute toxicities: 1 dysphagia and 1 mucositis. 5 (5.9%)Grade 3 or higher late toxicities 17.3 0 3-5 70.2 (39.6-134.4) 24 (3.1-252.6) 36
30 (18-40)
Number of Patients With Concurrent Cetuximab (%) SABR Median Fraction (Range) SABR Median Dose (Gy) (Range) Median Time From RT to SABR, Months (Range) Study No. of Prior RT Patients Median Dose (Gy) (Range)
Table 2 Retrospective Studies That Reviewed Reirradiation of Locally rSCCHN With SABR
Median Toxicities Followup (Months)
Toxicities following stereotactic ablative radiotherapy treatment
115 Patients were treated to a median dose of 30 Gy (range: 1641 Gy) in a median of 5 fractions (range: 3-5). The median prior RT dose was 68 Gy (range: 32.4-120.2 Gy), with all but 3 lesions treated with a minimum of 50 Gy. A total of 22 (25.9%) patients received cetuximab. Of all, 2 patients developed Grade 3 acute toxicities (1 dysphagia and 1 mucositis). A total of 5 patients developed Grade 3 or higher late toxicities. Table 1 summarized prior RT dose, SABR regimen and toxicities for all the prospective studies. Table 2 summarized similar parameters for all the retrospective studies. From comparing the results of prospective studies, it was clear that our institution had a low rate of Grade 3 toxicities (12%) while delivering a high radiation dose. More specifically, the median SABR radiation dose for both Heron et al and Vargo et al was 40 Gy, higher than that of Comet et al, Iwata et al, and Lartigau et al, but our Grade 3 toxicity rate in Heron et al was 0% and in Vargo et al was 12% (acute and late toxicities combined), much lower than 23% in Iwata et al and 32% in Lartigau et al. Retrospective studies showed similar results. Although the median SABR radiation dose (40 Gy) at our institution was much higher than those of other institutions (30 Gy for all of them), our toxicity rate was still comparable to that of other institutions. In summary, both prospective clinical trials and retrospective studies suggested that the technique at our institution allowed a delivery of relatively high radiation dose with SABR while maintaining a reasonable toxicity rate. In clinical practice, each retreatment depends heavily on the prior treatment and every patient should be treated as a special case. In general, based on our institution's experience, for patients who receive concurrent cetuximab with a median prior RT dose of 70 Gy, a total dose of 40-44 Gy delivered in 5 fractions on alternating days over 1-2 weeks is recommended. With this regimen, we expect 10%-20% Grade 2 or 3 toxicities and occasional Grade 4 toxicities, depending on previous RT dose, time interval from initial RT to SABR, tumor volume, SABR dose, number of treatment fraction and treatment duration.
Composite Plans and Application of NTCP Dose-Response Model Ideally, retreatment plans would be based on composite treatment plans including the original RT plan, the time interval from initial RT to SABR, and the reirradiation SABR plan.24-26 In practice, this is challenging because of limited access of complete previous treatment plans and the complexities of fusing the original plan with the retreatment plan as well as converting and summing the doses. The article by Leung et al24 presented an ideal approach but only included 1 case study, and NTCP estimates were based on parameters from prior studies, which may or may not be applicable. The articles by Nelson et al25 and Ma et al26 also provided ideal algorithms for computing composite doses and obtaining NTCP estimates, but these were for spinal cord retreatments instead of head and neck. Again, not much clinical data was explicitly provided. These 3 articles lay an important foundation, but clearly much more work is needed to fully understand head and neck retreatment planning.
K. Quan et al
116
Although SABR with concomitant cetuximab is generally well tolerated, Grade 2 or 3 toxicities, especially mucositis and dysphagia, are not uncommon in reirradiated patients. Therefore, we conducted a pilot study, applying an NTCP doseresponse model to estimate the probability of toxicities based on composite DVHs in SABR reirradiated locally rSCCHN.
normal tissue receiving doses more than 110% of the prescribed dose. The cumulative radiation dose to the spinal cord was not supposed to exceed 50 Gy at the equivalent dose of 2 Gy per fraction. Other normal tissues were constrained as much as possible provided that the target volume was not compromised. PTV volumes were often reduced from the skin to account for the dose buildup region in plan optimization. In cases where the GTV was extended to include critical organs based on tumor involvement, the treating radiation oncologist chose to reduce the PTV expansion in proximity to critical organs.
Patient Population
Treatment Characteristics and Modeling
A retrospective review of the medical records of 18 patients with previously irradiated, local rSCCHN treated at our institution between January 2008 and December 2011 was conducted after institutional review board approval. All patients were diagnosed with rSCCHN with disease locoregionally confined and previous irradiation dose Z60 Gy. All tumors were deemed unresectable by a multidisciplinary tumor board. The median age was 64 years (range: 45-87 years) with 13 (72%) males and 5 (28%) females. All patients received concomitant cetuximab. Of all, 11 (61%) patients received surgery and 10 (56%) patients received chemotherapy. The median previous radiation dose was 70 Gy (range: 63-118.2 Gy). The median time from prior radiation therapy to SABR was 11 months (range: 3-39 months). The detailed characteristics of patients and previous treatments are listed in Table 3.
All prescribed dose was delivered in 5 fractions. The median prescribed dose was 40 Gy (range: 40-44 Gy) with a median isodose line of 80% (range: 78%-93%). The median maximum radiation dose was 50 Gy (range: 47.3-55 Gy). The median number of days over which SABR was delivered was 9 days (range: 8-20 days). The median tumor volume was 24.4 ml (range: 4.4-75.7 ml). Out of these 18 patients, 9 (50%) were treated with Cyberknife; 6 (33.3%) were treated with Varian Trilogy; 3 (16.7%) were treated with Truebeam STX. The equivalent dose for 2 Gy per fraction (EQD2), assuming an α/β ratio of 3 Gy for normal organ effects, was calculated for each patient according to the formula EQD2 (Gy) ¼ total dose X (1 þ [dose per fraction]/3)/(1 þ 2/3). Both the IMRT plans and the SABR plans were converted voxel-by-voxel into EQD2 using an in-house digital imaging and communications in medicine (DICOM) software before any other analysis. For simplicity reasons, no factor was applied to account for the interval between the original RT and the SABR; the 2 EQD2 doses were simply added voxel-by-voxel. The EQD2, assuming an α/β ratio of 10 Gy for tumor effects, was also calculated for each patient's prescription dose in Table 2 according to the formula EQD2 (Gy) ¼ total dose X (1 þ [dose per fraction]/10)/(1 þ 2/10). The median prescribed SABR EQD2 was 88 Gy (range: 88-103.8 Gy) for α/β ¼ 3 Gy and was 60 Gy (range: 60-68.9 Gy) for α/β ¼ 10 Gy, respectively. The detailed information was listed in Table 4. Composite dose distributions of the EQD2 doses were generated via deformable registration in the VelocityAI software (Varian Inc, Palo Alto, CA). The composite DVHs were saved from VelocityAI and loaded into the DVH Evaluator software (DiversiLabs LLC, Huntingdon Valley, PA) for maximum likelihood parameter fitting of NTCP model parameters TD50 (v) and normalized slope γ50 of the exponential form of the logistic dose response model.28,29
Estimation of Toxicities in SABR Reirradiated Locally rSCCHN With Composite Dose NTCP Model: A Pilot Study
Treatment Technique Treatment started with a loading dose of cetuximab, 400 mg/ m2, intravenous (IV) infusion over 120 minutes on day -7 and then 250 mg/m2 on days 0 and þ8 during the SABR course. Patients were pre-medicated with either IV diphenhydramine hydrochloride, 50 mg (or an equivalent antihistamine) or dexamethasone, 8 mg orally or IV 30-60 minutes before each dose of cetuximab. An IV contrast-enhanced 18F-labeled fluorodeoxyglucose positron emission tomography-computed tomography scan was done for each patient before SABR for treatment planning. The images were reconstructed with a 1.25 mm slice thickness. During positron emission tomography-computed tomography imaging, the patient was immobilized with a thermoplastic mask. There was a 3-5 mm planning target volume (PTV) expansion on the gross tumor volume (GTV). Tumors with GTV less than 25 cm3 received 8.0 Gy per fraction for 5 fractions with a total radiation dose of 40 Gy; tumors with GTV more than 25 cm3 received 8.8 Gy per fraction for 5 fractions with a total radiation dose of 44 Gy.27 SABR was delivered on alternating days over 1-2 weeks. The prescription isodose lines were chosen to encompass at least 95% of the PTV with no more than 20% of PTV receiving doses 4110% of the prescribed dose, no more than 2% of any PTV receiving o93% of the prescribed dose and no more than 5% of any
NTCPðDv Þ ¼
e4g50 ðDv =TD50 ðvÞ1Þ 1 þ e4g50 ðDv =TD50 ðvÞ1Þ
ð1Þ
Toxicity Toxicity was graded based on the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events
Toxicities following stereotactic ablative radiotherapy treatment
117
Table 3 Patient and previous treatment characteristics Patient No.
Age at Treatment
Sex Initial Stage
1 2 3 4 5
58 72 67 61 85
M M M M M
6 7 8 9 10 11 12 13 14
49 65 64 47 75 64 81 58 62
M M F M F F M M F
15
57
M
16 17
87 45
F M
18
71
M
Site of Recurrence
T2N2M0 T3N0M0 T2N2cM0 T4aN0M0 T2N0M0
RPLN BOT Left level II LN Right RPLN Posterior pharyngeal wall T3N1M0 Nasopharynx T3N0M0 BOT T1N1M0 Left level II LN T3N2bM0 Parotid bed T3N2aM0 Left BOT T4aN2cM0 FOM T2N0M0 Oral tongue T2N2bM0 Right pterygoid T3N1M0 Right lateral pharyngeal wall T3N0M0 Posterior margin of reconstruction T4N0M0 Anterior maxilla T4N2cM0 Perimandibular site T2N2bM0 Right neck
Previous Surgery (Yes ¼ 1, No ¼ 0)
Previous Chemo (Yes ¼ 1, No ¼ 0)
Prior RT Dose (Gy)
Time From Prior RT to SABR (months)
1 0 1 1 1
0 0 0 1 1
70.2 70 72 118.2 70.2
12 9 17 4 31
0 0 1 1 1 0 0 0 0
0 1 1 1 1 0 0 0 0
70 73.8 65.1 70.2 70 70 70.2 66 70.2
31 13 9 10 33 8 6 23 9
1
1
70
39
1 1
1 1
63 66
32 6
1
1
66
3
Abbreviations: chemo, chemotherapy; RPLN, retropharyngeal lymph node; LN, lymph node; BOT, base of tongue; FOM, floor of mouth.
(CTCAE) v3.0. Only Grade 2 or greater toxicity was included for analysis. No Grade 4 or greater toxicity was observed in any of the 18 patients. The detailed information is provided in Table 5. In addition to the endpoints listed in Table 5, we also
monitored edema, trismus, hearing loss, hoarseness, nausea, vomiting, fibrosis, alopecia, and osteonecrosis. However, none of those toxicities occurred in our patients. Most notably, 1 (6%) patient had Grade 3 mucositis. Of all, 4 (22%) patients
Table 4 Characteristics of SABR Treatment Patient Prescribed Cyberknife ¼ 1, No. Dose (Gy) Triology ¼ 2, TrueBeam ¼ 3
Max Dose (Gy)
Period Over Isodose Tumor Which Delivered Line (%) Volume (days) (ml)
EQD2 (α/ EQD2 (α/ β ¼ 3) β¼10) (Gy) (Gy)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
50.0 48.8 50.0 50.0 50.0 50.0 51.3 55.0 51.0 50.0 47.3 50.0 50.0 48.9 51.2 50.0 52.4 48.9
9 8 9 8 13 20 9 14 9 9 9 9 9 9 9 12 12 12
88 88 88 88 88 88 88 103.8 91.1 88 103.8 88 88 103.8 103.8 88 103.8 103.8
40 40 40 40 40 40 40 44 40.8 40 44 40 40 44 44 40 44 44
1 2 1 2 1 1 1 2 1 2 3 1 1 2 3 1 2 3
80 82 80 80 80 80 78 80 80 80 93 80 80 90 86 80 84 90
21.6 65.6 7.1 16.1 11.9 8.1 64.6 57 19.3 13.6 42 28.6 22.7 4.4 75.7 52 26 39.6
60 60 60 60 60 60 60 68.9 61.7 60 68.9 60 60 68.9 68.9 60 68.9 68.9
Abbreviations: max, maximum; BED3, the biologically effective dose assuming an α/β ratio of 3 Gy for normal organ effects; BED10, the biologically effective dose assuming an α/β ratio of 10 Gy for tumor effects.
K. Quan et al
1 2
1
1
1
1
1
1
1
1
1
1 3
Figure Logistic model of median composite dose to oral mucosa. Abbreviations: AE, adverse event; ML, maximum likelihood. (Color version of figure is available online.)
1
1
had Grade 2 mucositis and 6 (33%) patients had Grade 1 mucositis.
2 2
2 1
3 1 1 1 2 2
1 2 1
1
1
2
1 2
2
NTCP Model Results Grade 2-3 mucositis correlated with oral mucosa composite DVH data, as shown in the Figure. Outcomes were better correlated with the median dose to oral mucosa than with doses for very small volumes. The fitted logistic model parameters for oral mucosa were TD50 (v) ¼ 190.8 Gy and g50 ¼ 0.3876. To keep the risk of Grade 2-3 mucositis below 25%, 30%, 33%, and 50%, the median composite EQD2 dose to the oral mucosa should be kept below 55.6 Gy, 86.5 Gy, 105.5 Gy, and 190.8 Gy, respectively. SABR dose tolerance limits are usually specified in terms of very small volumes like 1 or 0.1 ml, but in this study the larger volumes correlated better with outcomes. It is important to note that clinically for these cases, only small volumes of critical structures were allowed to exceed 20 Gy in 5 fractions. These tight constraints to small volumes probably are among the main factors that limited the occurrence of severe complications, but they also prevent us from seeing the full characteristics of dose-response for small volumes, although the large volume dose-response still provided some important insights.
2
1
1
1
Conclusion
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Patient Xerostomia Dysgeusia Mucositis Dysphagia Odynophagia Dysphonia Pain Fatigue Rash No. (Erbitux)
Table 5 Toxicity Results
Skin (Hyperpigmentation þ Fibrosis þ Alopecia)
2
Stomatitis
118
SABR with concomitant cetuximab has been shown to be an effective treatment for previously irradiated, locally rSCCHN. It is associated with good local control, OS and reasonable toxicities. However, Grade 2 or 3 toxicities, especially mucositis and dysphagia, are not uncommon. We conducted a pilot study that applied the NTCP dose-response model to estimate the probability of toxicities in those patients based on composite DVHs. The results showed that the high radiation dose to small volumes has been sufficiently constrained to
Toxicities following stereotactic ablative radiotherapy treatment avoid complications, to the extent that complications in this study correlated better to the median dose. Each retreatment depends heavily on the prior treatment and every patient is a special case. Based on our institution's experience and previously published studies, for patients who receive concurrent cetuximab with a median prior RT dose of 70 Gy, a total dose of 40-44 Gy in 5 fractions on alternating days over 12 weeks is recommended. Although we currently have limited data on the treatment of rSCCHN with SABR without cetuximab, the above dose limits are also reasonable starting points for treatment with SABR without any concurrent biological therapy or chemotherapy. Our DVH model can potentially estimate the risk of common toxicities such as mucositis based on median composite radiation dose to critical organs such as oral mucosa, which may allow for better planning to decrease the possibilities of toxicities. Although we understand mucositis is not the most significant toxicity, it was the only endpoint that occurred enough for us to construct the DVH model. Currently we are continuing to accumulate more patients with longer follow-ups. Once we have a much larger patient number, we should be able to construct DVH models for other more interesting toxicities such as dysphagia, dysphonia, trismus and even osteonecrosis, based on the median composite radiation dose to the relevant critical organs.
119
11.
12.
13.
14.
15.
16.
17.
18.
19.
References 1. Forastiere AA, Goepfert H, Maor M, et al: Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 349:2091-2098, 2003 2. Cooper JS, Pajak TF, Forastiere AA, et al: Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 350:1937-1944, 2004 3. Fakhry C, Zhang Q, Nguyen-Tan PF, et al: Human papillomavirus and overall survival after progression of oropharyngeal squamous cell carcinoma. J Clin Oncol 32:3365-3373, 2014 4. Dakubo GD, Jakupciak JP, Birch-Machin MA, et al: Clinical implications and utility of field cancerization. Cancer Cell Int 7:2, 2007 5. Cooper JS, Pajak TF, Rubin P, et al: Second malignancies in patients who have head and neck cancer: Incidence, effect on survival and implications based on the RTOG experience. Int J Radiat Oncol Biol Phys 17:449-456, 1989 6. Coatesworth AP, Tsikoudas A, MacLennan K: The cause of death in patients with head and neck squamous cell carcinoma. J Laryngol Otol 116:269-271, 2002 7. Clavel M, Vermorken JB, Cognetti F, et al: Randomized comparison of cisplatin, methotrexate, bleomycin and vincristine (CABO) versus cisplatin and 5-fluorouracil (CF) versus cisplatin (C) in recurrent or metastatic squamous cell carcinoma of the head and neck. A phase III study of the EORTC Head and Neck Cancer Cooperative Group. Ann Oncol 5:521-526, 1994 8. Argiris A, Li Y, Forastiere A: Prognostic factors and long-term survivorship in patients with recurrent or metastatic carcinoma of the head and neck. Cancer 101:2222-2229, 2004 9. Sulman EP, Schwartz DL, Le TT, et al: IMRT reirradiation of head and neck cancer-disease control and morbidity outcomes. Int J Radiat Oncol Biol Phys 73:399-409, 2009 10. Salama JK, Vokes EE, Chmura SJ, et al: Long-term outcome of concurrent chemotherapy and reirradiation for recurrent and second primary
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
head-and-neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 64:382-391, 2006 Bonner JA, Harari PM, Giralt J, et al: Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 354:567-578, 2006 Heron DE, Rwigema JC, Gibson MK, et al: Concurrent cetuximab with stereotactic body radiotherapy for recurrent squamous cell carcinoma of the head and neck: A single institution matched case-control study. Am J Clin Oncol 34:165-172, 2011 John A, Vargo RLF, Ohr James, et al: Clump: A prospective phase 2 trial of reirradiation with stereotactic body radiation therapy plus cetuximab in patients with previously irradiated recurrent squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 91:480-488, 2015 Heron DE, Ferris RL, Karamouzis M, et al: Stereotactic body radiotherapy for recurrent squamous cell carcinoma of the head and neck: Results of a phase I dose-escalation trial. Int J Radiat Oncol Biol Phys 75:1493-1500, 2009 Comet B, Kramar A, Faivre-Pierret M, et al: Salvage stereotactic reirradiation with or without cetuximab for locally recurrent head-and-neck cancer: A feasibility study. Int J Radiat Oncol Biol Phys 84:203-209, 2012 Iwata H, Tatewaki K, Inoue M, et al: Salvage stereotactic reirradiation using the CyberKnife for the local recurrence of nasal or paranasal carcinoma. Radiother Oncol 104:355-360, 2012 Lartigau EF, Tresch E, Thariat J, et al: Multi institutional phase II study of concomitant stereotactic reirradiation and cetuximab for recurrent head and neck cancer. Radiother Oncol 109:281-285, 2013 Vargo JA, Ferris RL, Ohr J, et al: A prospective phase 2 trial of reirradiation with stereotactic body radiation therapy plus cetuximab in patients with previously irradiated recurrent squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 91:480-488, 2015 Roh KW, Jang JS, Kim MS, et al: Fractionated stereotactic radiotherapy as reirradiation for locally recurrent head and neck cancer. Int J Radiat Oncol Biol Phys 74:1348-1355, 2009 Unger KR, Lominska CE, Deeken JF, et al: Fractionated stereotactic radiosurgery for reirradiation of head-and-neck cancer. Int J Radiat Oncol Biol Phys 77:1411-1419, 2010 Cengiz M, Ozyigit G, Yazici G, et al: Salvage reirradiaton with stereotactic body radiotherapy for locally recurrent head-and-neck tumors. Int J Radiat Oncol Biol Phys 81:104-109, 2011 Vargo JA, Heron DE, Ferris RL, et al: Examining tumor control and toxicity after stereotactic body radiotherapy in locally recurrent previously irradiated head and neck cancers: Implications of treatment duration and tumor volume. Head Neck 36:1349-1355, 2014 Kress MA, Sen N, Unger KR, et al: Safety and efficacy of hypofractionated stereotactic body reirradiation in head and neck cancer: Long-term followup of a large series. Head Neck 37(10):1403-1409, 2015 Leung LH, Wong WK, Cheng AC, et al: A new approach to computing normal tissue complication probability of an intensity-modulated radiotherapy treatment with stereotactic radiotherapy boost of nasopharyngeal carcinoma: A case study. Med Dosim 36:138-144, 2011 Nelson JW, Yoo DS, Sampson JH, et al: Stereotactic body radiotherapy for lesions of the spine and paraspinal regions. Int J Radiat Oncol Biol Phys 73:1369-1375, 2009 Ma L, Kirby N, Korol R, et al: Assessing small-volume spinal cord dose for repeat spinal stereotactic body radiotherapy treatments. Phys Med Biol 57:7843-7851, 2012 Rwigema JC, Heron DE, Ferris RL, et al: The impact of tumor volume and radiotherapy dose on outcome in previously irradiated recurrent squamous cell carcinoma of the head and neck treated with stereotactic body radiation therapy. Am J Clin Oncol 34:372-379, 2011 Okunieff P, Morgan D, Niemierko A, et al: Radiation dose-response of human tumors. Int J Radiat Oncol Biol Phys 32:1227-1237, 1995 Moiseenko V, Song WY, Mell LK, et al: A comparison of dose-response characteristics of four NTCP models using outcomes of radiation-induced optic neuropathy and retinopathy. Radiat Oncol 6:61, 2011
T
he locoregional recurrence rate in squamous cell carcinoma of the head and neck (SCCHN) following definitive and adjuvant chemoradiation therapy remains substantial at approximately 20%-30%,1,2 despite major improvements in the multimodality treatment of SCCHN. Radiation Therapy Oncology Group protocols reported locoregional recurrences as first failure sites in 50%-60% of both human papillomavirus-positive and human papillomavirus-negative subsets.3 An additional problem is the development of second primary head and neck cancers. The concept of field cancerization was first coined by Slaughter in 1953, who noticed the presence of histologically abnormal tissue surrounding oral
Department of Radiation Oncology, University of Pittsburgh Cancer Institute, Pittsburgh, PA. Conflicts of interest: none. Address reprint requests to Dwight E. Heron, MD, FACRO, Department of Radiation Oncology, University of Pittsburgh Cancer Institute, 5230 Centre Ave, Pittsburgh, PA 15232. E-mail: [email protected]
112
http://dx.doi.org/10.1016/j.semradonc.2015.11.007 1053-4296//& 2016 Elsevier Inc. All rights reserved.
squamous cell carcinoma. He proposed that the entire mucosa undergoes histological changes when exposed to environmental carcinogens such as tobacco and alcohol and is therefore susceptible to developing multiple primary malignancies.4 Based on this concept, the risk of a second primary malignancy for successfully treated SCCHN associated with tobacco and alcohol use is at least 1% per year.5 As locoregional progression is the most common cause of death,6 achieving local control in previously irradiated, locally recurrent disease may have a potentially significant effect on survival. Good local control also improves quality of life as uncontrolled tumor growth is associated with significant pain, bleeding, and impairment of essential functions such as swallowing, speaking, and breathing. For patients with resectable SCCHN recurrence, surgical salvage is the standard treatment.6 Unfortunately, many patients with previously irradiated locally recurrent SCCHN (rSCCHN) have tumors that are unresectable and reirradiation with or without chemotherapy becomes the only potentially
Toxicities following stereotactic ablative radiotherapy treatment curative salvage option. However, prospective randomized clinical trials studying conventional reirradiation techniques with chemotherapy reported significant toxicities with Grade 3 or higher acute toxicity rates of 64%-78% and Grade 3 or higher late toxicity rates of 22%-38%.7,8 Even modern intensity-modulated radiation therapy (IMRT) is associated with high rates of mucosal and soft tissue toxicities.9,10 Stereotactic ablative radiotherapy (SABR), owing to its great conformity and precision in dose delivery, has been associated with fewer normal tissue toxicities than conventional radiotherapy and IMRT. Furthermore, SABR requires much shorter treatment time (around 1-2 weeks) compared with IMRT, which can take 6-7 weeks. Since 2003, patients with previously irradiated rSCCHN have been treated on protocols with SABR ⫾ cetuximab at our institution.11-13 Although the treatment was in general well tolerated, we did observe some Grade 2-3 toxicities and occasional Grade 4 toxicities. The purpose of this review is to evaluate the toxicities of SABR in treating locally rSCCHN with previous irradiation and report results of a pilot study that applies the normal tissue complication probability (NTCP) dose-response model to study the dose-volume relationship to toxicities in previously irradiated, locally rSCCHN treated with SABR.
Toxicities of SABR in Reirradiating Locally rSCCHN Prospective Clinical Trials Several randomized, prospective clinical trials have investigated the use of SABR in reirradiating locally rSCCHN, as summarized in Table 1. Heron et al14 reported the first Phase I dose-escalation trial of reirradiating with SABR. Overall, 25 patients with locally rSCCHN, median tumor volume ¼ 44.8 (4.2-217) cm3, were retreated using SABR in 5 fractions with a total dose of 25-44 Gy, in 5 dose groups. The mean prior dose in each escalated dose group ranged from 66.0-69.6 Gy in 3036 fractions.14 Nobody received concurrent cetuximab or chemotherapy. The median time to disease progression after completion of SABR was 4 months and the median overall survival (OS) after completion of SABR was 6 months. No Grade 3 or 4 or dose-limiting toxicities occurred. In all, 2 patients experienced Grade 1 mucositis, 1 patient with Grade 2 dysphagia and 1 patient with Grade 1 hyperpigmentation. Selfreported quality of life was not significantly affected by SABR treatment. In 2012, Comet et al15 conducted a prospective study that included 40 patients with inoperable or new primary head and neck cancer in a previously irradiated area. Overall, 50% of these tumors were squamous cell carcinoma. The median prior RT dose was 66 Gy and median tumor diameter was 29 mm. Overall, 70% had previous surgery and 57% had received chemotherapy during their initial treatment. SABR was delivered in 6 fractions to the 85% isodose line with a total dose of 36 Gy. In this study, 15 patients received concurrent cetuximab and 1 received concurrent cisplatin. The median progression-free survival was 8.8 months (95% CI: 5.2-11.7
113 months) and the median OS was 13.6 months. The median follow-up was 25.6 months. No Grade 4 toxicity was observed. Of all, 4 (10.3%) patients developed Grade 3 toxicities including mucositis, dysphagia, induration and fibrosis, 3 of whom received concurrent cetuximab. In all, 8 (20.6%) patients experienced Grade 2 toxicities and 10 (25.6%) experienced Grade 1 toxicities. Iwata et al16 reported a prospective protocol-based study that included 51 patients with previously irradiated recurrent nasal or paranasal carcinoma. Most patients were reirradiated with a total dose of 30 Gy in 3 fractions or 35 Gy in 5 fractions. The median prior RT dose was 60 Gy (range: 40-70 Gy). The median interval between initial radiation and reirradiation was 18 months (range: 2.5-132 months). The median time-to-local progression after completion of SABR was 9.5 months and the median OS after completion of SABR was 14.5 months. Grade 3 or higher adverse events were observed in 23% of patients. A total of 2 patients developed Grade 4 dermatitis and 1 patient developed Grade 4 soft tissue necrosis. Of all, 2 brain necrosis patients presented with clinical symptoms. Lartigau et al17 reported a multiinstitution Phase II study that included 56 patients with previously irradiated, locally rSCCHN. SABR was delivered in 6 fractions to the 85% isodose line with a total dose of 36 Gy. All patients received concomitant cetuximab. The median prior RT dose was not reported. The median progression-free survival after completion of SABR was 7.1 months (95% CI: 5.5-8.9 months) and the median OS after completion of SABR was 11.8 months with a 1-year OS rate of 47.5% (95% CI: 30.8%-62.4%). The median follow-up was 11.4 months. A total of 18 patients developed Grade 3 toxicities including mucositis, dysphagia, induration, and fibrosis. More recently, Vargo et al18 reported a prospective Phase II trial of 50 patients with inoperable locally rSCCHN with a previous RT dose Z60 Gy and tumor volumes of 3.6-209.2 (median ¼ 36.5) cm3. The median time from initial radiation to reirradiation was 18 months (range: 3-423 months). The median prior RT dose was 70 Gy (range: 52.5-118.2 Gy). All patients received concomitant cetuximab. SABR delivered a total dose of 40-44 Gy in 5 fractions on alternating days over 12 weeks. The median OS was 10 months (95% CI: 7-16 months) with a 1-year local progression-free survival rate of 60% (95% CI: 44%-75%). The median follow-up was 18 months (range: 10-70 months). Of all, 6% of patients developed Grade 3 acute toxicities including 1 mucositis, 1 dysphagia and 1 skin rash. A total of 6% of patients developed Grade 3 late toxicities including 1 dysphagia (45 cm3 tonsil tumor), and 2 aerodigestive fistulae (45 cm3 stoma/submental and 135 cm3 base of tongue). There was no Grade 4 or greater toxicity. In summary, SABR is a feasible option to re-irradiate locally rSCCHN. It is associated with a reasonable local control rate, OS and toxicity profile. However, Grade 2 or 3 toxicities are not uncommon after treatment. Most commonly seen toxicities include mucositis, dysphagia, and dermatitis.
K. Quan et al
18 50 (100%) 5 70 (52.5-118.2) 18 (3-423) 50 Vargo et al18
40-44
11.4 56 (100%) 6 36 38 Not reported Lartigau et al17 56
21 0 1-5 35 (20-41.5) 60 (40-70)
18 (2.5-132) 51 Iwata et al16
66
31.6 (7.9-263.4)
36 40 Comet et al15
6
15 (37.5%)
25.6
2 (%) Grade 1 mucositis, 1 (4%) Grade 2 dysphagia, and 1 (4%) Grade 1 hyperpigmentation No Grade 4 toxicity, 4 (10.3%) Grade 3 toxicities, 8 (20.6%) Grade 2 toxicities, and 10 (25.6%) Grade 1 toxicities Grade 3 or higher adverse events in 23% of patients, 2 (3.9%) Grade 4 dermatitis, and 1 (1.96%) Grade 4 soft tissue necrosis 18 (32.1%) Grade 3 toxicities: mucositis, dysphagia, induration and fibrosis 3 (6%) Grade 3 acute toxicities: 1 mucositis, 1 dysphagia and 1 skin rash. 3 Grade 3 (6%) late toxicities: 1 dysphagia, and 2 aerodigestive fistulas. No Grade 4 or greater toxicity Not Reported 0 5 64.7
Not Reported
40 (25-44) 25 Heron et al14
SABR Median Fraction (Range) SABR Median Time No. of Prior RT Patients Median Dose From RT to SABR, Median Dose (Gy) (Gy) (Range) Months (Range) (Range) Study
Table 1 Prospective Clinical Trials That Investigated Reirradiating Locally rSCCHN With SABR
Toxicities Median No. of Patients With Concurrent Follow-up (Months) Cetuximab (%)
114
Retrospective Studies Multiple retrospective studies investigated toxicities of SABR in treating locally rSCCHN with previous irradiation, as summarized in Table 2. Roh et al19 reported a study of 36 patients reirradiated with SABR for locally rSCCHN. A total dose of 1840 Gy was delivered in 3-5 fractions to the 65%-85% isodose line. The median prior RT dose was 70.2 Gy (range: 39.6134.4 Gy). The median follow-up was 17.3 months. Grade 3 acute toxicities were observed in 13 (36.1%) patients and late complications (1 bone necrosis and 2 soft tissue necrosis) were observed in 3 (8.3%) patients. Unger et al20 reported a study of 65 patients including 38 treated definitively and 27 treated for metastatic diseases. A total of 11 (17%) patients received concurrent cetuximab. The median initial RT dose was 66.6 Gy (range: 32.4-120.2 Gy). SABR delivered a median dose of 30 Gy (range: 21-35 Gy) in 2-5 fractions. The median OS for all patients was 12 months. No patients experienced Grade 4 acute toxicity. A total of 19 patients experienced Grade 1-3 acute toxicities, including mucositis, dermatitis, and nausea. A total of 6 (9%) patients experienced Grade 4 toxicities, including arterial bleeding, soft tissue necrosis, fistula formation and dysphagia. A single patient died from treatment-related complications. Cengiz et al21 reported 46 locally recurrent head-and-neck tumors irradiated with SABR. The most common histopathology was epidermoid carcinoma. The median prior RT dose was 61 Gy (range: 30-70 Gy). SABR delivered a median dose of 30 Gy (range: 18-35 Gy) in a median of 5 (range: 1-5) fractions. The median OS after completion of SABR was 11.9 months (range: 11.4-17.4 months). A single (2.2%) patient developed Grade 2 acute mucositis. Of all, 2 (4.4%) patients experienced Grade 3 acute dermatitis and mucositis. Overall, 6 (13.3%) patients developed long-term complications. A single patient had soft tissue necrosis and 1 patient developed mandibular necrosis. Of all, 2 (4.4%) patients developed Grade 3 dysphagia. Of all, 2 (4.4%) patients developed Grade 2 dysphagia. Vargo et al22 retrospectively reviewed 132 patients with previously irradiated, locally rSCCHN with median tumor volume of 30.9 (4.4-192.4) cm3. SABR was delivered in 5 fractions with a total dose of 44-50 Gy. A total of 55% of patients received concomitant cetuximab. The median prior RT dose was 68 Gy (range: 20.4-140 Gy) at a reirradiation interval of 23 months (range: 2-423 months). A total of 3 patients developed Grade 3 acute mucositis and 1 developed Grade 4 acute mucositis. A single patient developed Grade 3 acute dysphagia and 2 patients developed Grade 4 acute dysphagia. Of all, 3 patients developed Grade 3 skin rash and 1 developed Grade 4 skin rash. At a median follow-up of 6 months (range: 0-55 months), treatment duration o14 days was associated with significantly improved recurrence-free survival (37% vs 8% 1-year, p ¼ 0.041), in univariate but not multivariate analysis at the expense of increased late toxicity (8% vs 0%, p ¼ 0.017). Tumor volume 425 ml was associated with significantly more acute toxicity (75% vs 66%, p ¼ 0.017) in univariate and multivariate analysis but had no effect on late toxicity. 23 More recently, Kress et al retrospectively reviewed 85 patients with previously irradiated SCCHN treated with SABR.
17.3 5 (3-5) Kress 85 et al23
68 (32.4-120.2)
32 (1-324)
30 (16-41)
22 (25.9%)
6 5 23 (2-423) 68 (20.4-140) 132
44 (35-50)
72 (55%)
7 0 5 (1-5) 30 (18-35)
Roh et al19 Unger et al20 Cengiz et al21 Vargo et al22
38 (3.8-306) 61 (30-70) 46
2-5 66.6 (32.4-120.2) 26 (1.6-318) 65
30 (21-35)
11 (17%)
12
13 (36%) Grade 3 acute toxicities and 3 (8.3%) late complications: 1 bone necrosis and 2 soft tissue necrosis 19 (29%) Grade 1-3 acute toxicities. 6 (9%) Grade 4 toxicities. 1 died from treatment-related complications 1 (2.2%) Grade 2 acute mucositis. 2 (4.4%) Grade 3 acute dermatitis and mucositis. 6 (13.3%) late complications 3 (2.3%) Grade 3 acute mucositis and 1 (0.75%) Grade 4 acute mucositis. 1 (0.75%) Grade 3 acute dysphagia and 2 (1.5%) Grade 4 acute dysphagia. 3 (2.3%) Grade 3 skin rash and 1 (0.75%) Grade 4 skin rash 2 (2.4%) Grade 3 acute toxicities: 1 dysphagia and 1 mucositis. 5 (5.9%)Grade 3 or higher late toxicities 17.3 0 3-5 70.2 (39.6-134.4) 24 (3.1-252.6) 36
30 (18-40)
Number of Patients With Concurrent Cetuximab (%) SABR Median Fraction (Range) SABR Median Dose (Gy) (Range) Median Time From RT to SABR, Months (Range) Study No. of Prior RT Patients Median Dose (Gy) (Range)
Table 2 Retrospective Studies That Reviewed Reirradiation of Locally rSCCHN With SABR
Median Toxicities Followup (Months)
Toxicities following stereotactic ablative radiotherapy treatment
115 Patients were treated to a median dose of 30 Gy (range: 1641 Gy) in a median of 5 fractions (range: 3-5). The median prior RT dose was 68 Gy (range: 32.4-120.2 Gy), with all but 3 lesions treated with a minimum of 50 Gy. A total of 22 (25.9%) patients received cetuximab. Of all, 2 patients developed Grade 3 acute toxicities (1 dysphagia and 1 mucositis). A total of 5 patients developed Grade 3 or higher late toxicities. Table 1 summarized prior RT dose, SABR regimen and toxicities for all the prospective studies. Table 2 summarized similar parameters for all the retrospective studies. From comparing the results of prospective studies, it was clear that our institution had a low rate of Grade 3 toxicities (12%) while delivering a high radiation dose. More specifically, the median SABR radiation dose for both Heron et al and Vargo et al was 40 Gy, higher than that of Comet et al, Iwata et al, and Lartigau et al, but our Grade 3 toxicity rate in Heron et al was 0% and in Vargo et al was 12% (acute and late toxicities combined), much lower than 23% in Iwata et al and 32% in Lartigau et al. Retrospective studies showed similar results. Although the median SABR radiation dose (40 Gy) at our institution was much higher than those of other institutions (30 Gy for all of them), our toxicity rate was still comparable to that of other institutions. In summary, both prospective clinical trials and retrospective studies suggested that the technique at our institution allowed a delivery of relatively high radiation dose with SABR while maintaining a reasonable toxicity rate. In clinical practice, each retreatment depends heavily on the prior treatment and every patient should be treated as a special case. In general, based on our institution's experience, for patients who receive concurrent cetuximab with a median prior RT dose of 70 Gy, a total dose of 40-44 Gy delivered in 5 fractions on alternating days over 1-2 weeks is recommended. With this regimen, we expect 10%-20% Grade 2 or 3 toxicities and occasional Grade 4 toxicities, depending on previous RT dose, time interval from initial RT to SABR, tumor volume, SABR dose, number of treatment fraction and treatment duration.
Composite Plans and Application of NTCP Dose-Response Model Ideally, retreatment plans would be based on composite treatment plans including the original RT plan, the time interval from initial RT to SABR, and the reirradiation SABR plan.24-26 In practice, this is challenging because of limited access of complete previous treatment plans and the complexities of fusing the original plan with the retreatment plan as well as converting and summing the doses. The article by Leung et al24 presented an ideal approach but only included 1 case study, and NTCP estimates were based on parameters from prior studies, which may or may not be applicable. The articles by Nelson et al25 and Ma et al26 also provided ideal algorithms for computing composite doses and obtaining NTCP estimates, but these were for spinal cord retreatments instead of head and neck. Again, not much clinical data was explicitly provided. These 3 articles lay an important foundation, but clearly much more work is needed to fully understand head and neck retreatment planning.
K. Quan et al
116
Although SABR with concomitant cetuximab is generally well tolerated, Grade 2 or 3 toxicities, especially mucositis and dysphagia, are not uncommon in reirradiated patients. Therefore, we conducted a pilot study, applying an NTCP doseresponse model to estimate the probability of toxicities based on composite DVHs in SABR reirradiated locally rSCCHN.
normal tissue receiving doses more than 110% of the prescribed dose. The cumulative radiation dose to the spinal cord was not supposed to exceed 50 Gy at the equivalent dose of 2 Gy per fraction. Other normal tissues were constrained as much as possible provided that the target volume was not compromised. PTV volumes were often reduced from the skin to account for the dose buildup region in plan optimization. In cases where the GTV was extended to include critical organs based on tumor involvement, the treating radiation oncologist chose to reduce the PTV expansion in proximity to critical organs.
Patient Population
Treatment Characteristics and Modeling
A retrospective review of the medical records of 18 patients with previously irradiated, local rSCCHN treated at our institution between January 2008 and December 2011 was conducted after institutional review board approval. All patients were diagnosed with rSCCHN with disease locoregionally confined and previous irradiation dose Z60 Gy. All tumors were deemed unresectable by a multidisciplinary tumor board. The median age was 64 years (range: 45-87 years) with 13 (72%) males and 5 (28%) females. All patients received concomitant cetuximab. Of all, 11 (61%) patients received surgery and 10 (56%) patients received chemotherapy. The median previous radiation dose was 70 Gy (range: 63-118.2 Gy). The median time from prior radiation therapy to SABR was 11 months (range: 3-39 months). The detailed characteristics of patients and previous treatments are listed in Table 3.
All prescribed dose was delivered in 5 fractions. The median prescribed dose was 40 Gy (range: 40-44 Gy) with a median isodose line of 80% (range: 78%-93%). The median maximum radiation dose was 50 Gy (range: 47.3-55 Gy). The median number of days over which SABR was delivered was 9 days (range: 8-20 days). The median tumor volume was 24.4 ml (range: 4.4-75.7 ml). Out of these 18 patients, 9 (50%) were treated with Cyberknife; 6 (33.3%) were treated with Varian Trilogy; 3 (16.7%) were treated with Truebeam STX. The equivalent dose for 2 Gy per fraction (EQD2), assuming an α/β ratio of 3 Gy for normal organ effects, was calculated for each patient according to the formula EQD2 (Gy) ¼ total dose X (1 þ [dose per fraction]/3)/(1 þ 2/3). Both the IMRT plans and the SABR plans were converted voxel-by-voxel into EQD2 using an in-house digital imaging and communications in medicine (DICOM) software before any other analysis. For simplicity reasons, no factor was applied to account for the interval between the original RT and the SABR; the 2 EQD2 doses were simply added voxel-by-voxel. The EQD2, assuming an α/β ratio of 10 Gy for tumor effects, was also calculated for each patient's prescription dose in Table 2 according to the formula EQD2 (Gy) ¼ total dose X (1 þ [dose per fraction]/10)/(1 þ 2/10). The median prescribed SABR EQD2 was 88 Gy (range: 88-103.8 Gy) for α/β ¼ 3 Gy and was 60 Gy (range: 60-68.9 Gy) for α/β ¼ 10 Gy, respectively. The detailed information was listed in Table 4. Composite dose distributions of the EQD2 doses were generated via deformable registration in the VelocityAI software (Varian Inc, Palo Alto, CA). The composite DVHs were saved from VelocityAI and loaded into the DVH Evaluator software (DiversiLabs LLC, Huntingdon Valley, PA) for maximum likelihood parameter fitting of NTCP model parameters TD50 (v) and normalized slope γ50 of the exponential form of the logistic dose response model.28,29
Estimation of Toxicities in SABR Reirradiated Locally rSCCHN With Composite Dose NTCP Model: A Pilot Study
Treatment Technique Treatment started with a loading dose of cetuximab, 400 mg/ m2, intravenous (IV) infusion over 120 minutes on day -7 and then 250 mg/m2 on days 0 and þ8 during the SABR course. Patients were pre-medicated with either IV diphenhydramine hydrochloride, 50 mg (or an equivalent antihistamine) or dexamethasone, 8 mg orally or IV 30-60 minutes before each dose of cetuximab. An IV contrast-enhanced 18F-labeled fluorodeoxyglucose positron emission tomography-computed tomography scan was done for each patient before SABR for treatment planning. The images were reconstructed with a 1.25 mm slice thickness. During positron emission tomography-computed tomography imaging, the patient was immobilized with a thermoplastic mask. There was a 3-5 mm planning target volume (PTV) expansion on the gross tumor volume (GTV). Tumors with GTV less than 25 cm3 received 8.0 Gy per fraction for 5 fractions with a total radiation dose of 40 Gy; tumors with GTV more than 25 cm3 received 8.8 Gy per fraction for 5 fractions with a total radiation dose of 44 Gy.27 SABR was delivered on alternating days over 1-2 weeks. The prescription isodose lines were chosen to encompass at least 95% of the PTV with no more than 20% of PTV receiving doses 4110% of the prescribed dose, no more than 2% of any PTV receiving o93% of the prescribed dose and no more than 5% of any
NTCPðDv Þ ¼
e4g50 ðDv =TD50 ðvÞ1Þ 1 þ e4g50 ðDv =TD50 ðvÞ1Þ
ð1Þ
Toxicity Toxicity was graded based on the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events
Toxicities following stereotactic ablative radiotherapy treatment
117
Table 3 Patient and previous treatment characteristics Patient No.
Age at Treatment
Sex Initial Stage
1 2 3 4 5
58 72 67 61 85
M M M M M
6 7 8 9 10 11 12 13 14
49 65 64 47 75 64 81 58 62
M M F M F F M M F
15
57
M
16 17
87 45
F M
18
71
M
Site of Recurrence
T2N2M0 T3N0M0 T2N2cM0 T4aN0M0 T2N0M0
RPLN BOT Left level II LN Right RPLN Posterior pharyngeal wall T3N1M0 Nasopharynx T3N0M0 BOT T1N1M0 Left level II LN T3N2bM0 Parotid bed T3N2aM0 Left BOT T4aN2cM0 FOM T2N0M0 Oral tongue T2N2bM0 Right pterygoid T3N1M0 Right lateral pharyngeal wall T3N0M0 Posterior margin of reconstruction T4N0M0 Anterior maxilla T4N2cM0 Perimandibular site T2N2bM0 Right neck
Previous Surgery (Yes ¼ 1, No ¼ 0)
Previous Chemo (Yes ¼ 1, No ¼ 0)
Prior RT Dose (Gy)
Time From Prior RT to SABR (months)
1 0 1 1 1
0 0 0 1 1
70.2 70 72 118.2 70.2
12 9 17 4 31
0 0 1 1 1 0 0 0 0
0 1 1 1 1 0 0 0 0
70 73.8 65.1 70.2 70 70 70.2 66 70.2
31 13 9 10 33 8 6 23 9
1
1
70
39
1 1
1 1
63 66
32 6
1
1
66
3
Abbreviations: chemo, chemotherapy; RPLN, retropharyngeal lymph node; LN, lymph node; BOT, base of tongue; FOM, floor of mouth.
(CTCAE) v3.0. Only Grade 2 or greater toxicity was included for analysis. No Grade 4 or greater toxicity was observed in any of the 18 patients. The detailed information is provided in Table 5. In addition to the endpoints listed in Table 5, we also
monitored edema, trismus, hearing loss, hoarseness, nausea, vomiting, fibrosis, alopecia, and osteonecrosis. However, none of those toxicities occurred in our patients. Most notably, 1 (6%) patient had Grade 3 mucositis. Of all, 4 (22%) patients
Table 4 Characteristics of SABR Treatment Patient Prescribed Cyberknife ¼ 1, No. Dose (Gy) Triology ¼ 2, TrueBeam ¼ 3
Max Dose (Gy)
Period Over Isodose Tumor Which Delivered Line (%) Volume (days) (ml)
EQD2 (α/ EQD2 (α/ β ¼ 3) β¼10) (Gy) (Gy)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
50.0 48.8 50.0 50.0 50.0 50.0 51.3 55.0 51.0 50.0 47.3 50.0 50.0 48.9 51.2 50.0 52.4 48.9
9 8 9 8 13 20 9 14 9 9 9 9 9 9 9 12 12 12
88 88 88 88 88 88 88 103.8 91.1 88 103.8 88 88 103.8 103.8 88 103.8 103.8
40 40 40 40 40 40 40 44 40.8 40 44 40 40 44 44 40 44 44
1 2 1 2 1 1 1 2 1 2 3 1 1 2 3 1 2 3
80 82 80 80 80 80 78 80 80 80 93 80 80 90 86 80 84 90
21.6 65.6 7.1 16.1 11.9 8.1 64.6 57 19.3 13.6 42 28.6 22.7 4.4 75.7 52 26 39.6
60 60 60 60 60 60 60 68.9 61.7 60 68.9 60 60 68.9 68.9 60 68.9 68.9
Abbreviations: max, maximum; BED3, the biologically effective dose assuming an α/β ratio of 3 Gy for normal organ effects; BED10, the biologically effective dose assuming an α/β ratio of 10 Gy for tumor effects.
K. Quan et al
1 2
1
1
1
1
1
1
1
1
1
1 3
Figure Logistic model of median composite dose to oral mucosa. Abbreviations: AE, adverse event; ML, maximum likelihood. (Color version of figure is available online.)
1
1
had Grade 2 mucositis and 6 (33%) patients had Grade 1 mucositis.
2 2
2 1
3 1 1 1 2 2
1 2 1
1
1
2
1 2
2
NTCP Model Results Grade 2-3 mucositis correlated with oral mucosa composite DVH data, as shown in the Figure. Outcomes were better correlated with the median dose to oral mucosa than with doses for very small volumes. The fitted logistic model parameters for oral mucosa were TD50 (v) ¼ 190.8 Gy and g50 ¼ 0.3876. To keep the risk of Grade 2-3 mucositis below 25%, 30%, 33%, and 50%, the median composite EQD2 dose to the oral mucosa should be kept below 55.6 Gy, 86.5 Gy, 105.5 Gy, and 190.8 Gy, respectively. SABR dose tolerance limits are usually specified in terms of very small volumes like 1 or 0.1 ml, but in this study the larger volumes correlated better with outcomes. It is important to note that clinically for these cases, only small volumes of critical structures were allowed to exceed 20 Gy in 5 fractions. These tight constraints to small volumes probably are among the main factors that limited the occurrence of severe complications, but they also prevent us from seeing the full characteristics of dose-response for small volumes, although the large volume dose-response still provided some important insights.
2
1
1
1
Conclusion
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Patient Xerostomia Dysgeusia Mucositis Dysphagia Odynophagia Dysphonia Pain Fatigue Rash No. (Erbitux)
Table 5 Toxicity Results
Skin (Hyperpigmentation þ Fibrosis þ Alopecia)
2
Stomatitis
118
SABR with concomitant cetuximab has been shown to be an effective treatment for previously irradiated, locally rSCCHN. It is associated with good local control, OS and reasonable toxicities. However, Grade 2 or 3 toxicities, especially mucositis and dysphagia, are not uncommon. We conducted a pilot study that applied the NTCP dose-response model to estimate the probability of toxicities in those patients based on composite DVHs. The results showed that the high radiation dose to small volumes has been sufficiently constrained to
Toxicities following stereotactic ablative radiotherapy treatment avoid complications, to the extent that complications in this study correlated better to the median dose. Each retreatment depends heavily on the prior treatment and every patient is a special case. Based on our institution's experience and previously published studies, for patients who receive concurrent cetuximab with a median prior RT dose of 70 Gy, a total dose of 40-44 Gy in 5 fractions on alternating days over 12 weeks is recommended. Although we currently have limited data on the treatment of rSCCHN with SABR without cetuximab, the above dose limits are also reasonable starting points for treatment with SABR without any concurrent biological therapy or chemotherapy. Our DVH model can potentially estimate the risk of common toxicities such as mucositis based on median composite radiation dose to critical organs such as oral mucosa, which may allow for better planning to decrease the possibilities of toxicities. Although we understand mucositis is not the most significant toxicity, it was the only endpoint that occurred enough for us to construct the DVH model. Currently we are continuing to accumulate more patients with longer follow-ups. Once we have a much larger patient number, we should be able to construct DVH models for other more interesting toxicities such as dysphagia, dysphonia, trismus and even osteonecrosis, based on the median composite radiation dose to the relevant critical organs.
119
11.
12.
13.
14.
15.
16.
17.
18.
19.
References 1. Forastiere AA, Goepfert H, Maor M, et al: Concurrent chemotherapy and radiotherapy for organ preservation in advanced laryngeal cancer. N Engl J Med 349:2091-2098, 2003 2. Cooper JS, Pajak TF, Forastiere AA, et al: Postoperative concurrent radiotherapy and chemotherapy for high-risk squamous-cell carcinoma of the head and neck. N Engl J Med 350:1937-1944, 2004 3. Fakhry C, Zhang Q, Nguyen-Tan PF, et al: Human papillomavirus and overall survival after progression of oropharyngeal squamous cell carcinoma. J Clin Oncol 32:3365-3373, 2014 4. Dakubo GD, Jakupciak JP, Birch-Machin MA, et al: Clinical implications and utility of field cancerization. Cancer Cell Int 7:2, 2007 5. Cooper JS, Pajak TF, Rubin P, et al: Second malignancies in patients who have head and neck cancer: Incidence, effect on survival and implications based on the RTOG experience. Int J Radiat Oncol Biol Phys 17:449-456, 1989 6. Coatesworth AP, Tsikoudas A, MacLennan K: The cause of death in patients with head and neck squamous cell carcinoma. J Laryngol Otol 116:269-271, 2002 7. Clavel M, Vermorken JB, Cognetti F, et al: Randomized comparison of cisplatin, methotrexate, bleomycin and vincristine (CABO) versus cisplatin and 5-fluorouracil (CF) versus cisplatin (C) in recurrent or metastatic squamous cell carcinoma of the head and neck. A phase III study of the EORTC Head and Neck Cancer Cooperative Group. Ann Oncol 5:521-526, 1994 8. Argiris A, Li Y, Forastiere A: Prognostic factors and long-term survivorship in patients with recurrent or metastatic carcinoma of the head and neck. Cancer 101:2222-2229, 2004 9. Sulman EP, Schwartz DL, Le TT, et al: IMRT reirradiation of head and neck cancer-disease control and morbidity outcomes. Int J Radiat Oncol Biol Phys 73:399-409, 2009 10. Salama JK, Vokes EE, Chmura SJ, et al: Long-term outcome of concurrent chemotherapy and reirradiation for recurrent and second primary
20.
21.
22.
23.
24.
25.
26.
27.
28. 29.
head-and-neck squamous cell carcinoma. Int J Radiat Oncol Biol Phys 64:382-391, 2006 Bonner JA, Harari PM, Giralt J, et al: Radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. N Engl J Med 354:567-578, 2006 Heron DE, Rwigema JC, Gibson MK, et al: Concurrent cetuximab with stereotactic body radiotherapy for recurrent squamous cell carcinoma of the head and neck: A single institution matched case-control study. Am J Clin Oncol 34:165-172, 2011 John A, Vargo RLF, Ohr James, et al: Clump: A prospective phase 2 trial of reirradiation with stereotactic body radiation therapy plus cetuximab in patients with previously irradiated recurrent squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 91:480-488, 2015 Heron DE, Ferris RL, Karamouzis M, et al: Stereotactic body radiotherapy for recurrent squamous cell carcinoma of the head and neck: Results of a phase I dose-escalation trial. Int J Radiat Oncol Biol Phys 75:1493-1500, 2009 Comet B, Kramar A, Faivre-Pierret M, et al: Salvage stereotactic reirradiation with or without cetuximab for locally recurrent head-and-neck cancer: A feasibility study. Int J Radiat Oncol Biol Phys 84:203-209, 2012 Iwata H, Tatewaki K, Inoue M, et al: Salvage stereotactic reirradiation using the CyberKnife for the local recurrence of nasal or paranasal carcinoma. Radiother Oncol 104:355-360, 2012 Lartigau EF, Tresch E, Thariat J, et al: Multi institutional phase II study of concomitant stereotactic reirradiation and cetuximab for recurrent head and neck cancer. Radiother Oncol 109:281-285, 2013 Vargo JA, Ferris RL, Ohr J, et al: A prospective phase 2 trial of reirradiation with stereotactic body radiation therapy plus cetuximab in patients with previously irradiated recurrent squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 91:480-488, 2015 Roh KW, Jang JS, Kim MS, et al: Fractionated stereotactic radiotherapy as reirradiation for locally recurrent head and neck cancer. Int J Radiat Oncol Biol Phys 74:1348-1355, 2009 Unger KR, Lominska CE, Deeken JF, et al: Fractionated stereotactic radiosurgery for reirradiation of head-and-neck cancer. Int J Radiat Oncol Biol Phys 77:1411-1419, 2010 Cengiz M, Ozyigit G, Yazici G, et al: Salvage reirradiaton with stereotactic body radiotherapy for locally recurrent head-and-neck tumors. Int J Radiat Oncol Biol Phys 81:104-109, 2011 Vargo JA, Heron DE, Ferris RL, et al: Examining tumor control and toxicity after stereotactic body radiotherapy in locally recurrent previously irradiated head and neck cancers: Implications of treatment duration and tumor volume. Head Neck 36:1349-1355, 2014 Kress MA, Sen N, Unger KR, et al: Safety and efficacy of hypofractionated stereotactic body reirradiation in head and neck cancer: Long-term followup of a large series. Head Neck 37(10):1403-1409, 2015 Leung LH, Wong WK, Cheng AC, et al: A new approach to computing normal tissue complication probability of an intensity-modulated radiotherapy treatment with stereotactic radiotherapy boost of nasopharyngeal carcinoma: A case study. Med Dosim 36:138-144, 2011 Nelson JW, Yoo DS, Sampson JH, et al: Stereotactic body radiotherapy for lesions of the spine and paraspinal regions. Int J Radiat Oncol Biol Phys 73:1369-1375, 2009 Ma L, Kirby N, Korol R, et al: Assessing small-volume spinal cord dose for repeat spinal stereotactic body radiotherapy treatments. Phys Med Biol 57:7843-7851, 2012 Rwigema JC, Heron DE, Ferris RL, et al: The impact of tumor volume and radiotherapy dose on outcome in previously irradiated recurrent squamous cell carcinoma of the head and neck treated with stereotactic body radiation therapy. Am J Clin Oncol 34:372-379, 2011 Okunieff P, Morgan D, Niemierko A, et al: Radiation dose-response of human tumors. Int J Radiat Oncol Biol Phys 32:1227-1237, 1995 Moiseenko V, Song WY, Mell LK, et al: A comparison of dose-response characteristics of four NTCP models using outcomes of radiation-induced optic neuropathy and retinopathy. Radiat Oncol 6:61, 2011