- Email: [email protected]
Endoluminal high-dose-rate brachytherapy for early stage and recurrent esophageal cancer in medically inoperable patients
Brachytherapy 12 (2013) 463e470
Endoluminal high-dose-rate brachytherapy for early stage and recurrent esophageal cancer in medically inoperable patients Michael R. Folkert1, Gil’ad N. Cohen2, Abraham J. Wu1, Hans Gerdes3, Mark A. Schattner3, Arnold J. Markowitz3, Emmy Ludwig3, David H. Ilson4, Manjit S. Bains5, Michael J. Zelefsky1, Karyn A. Goodman1,* 1
Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 2 Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY 3 Department of Gastroenterology, Memorial Sloan-Kettering Cancer Center, New York, NY 4 Department of Medical Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 5 Department of Thoracic Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY
ABSTRACT
PURPOSE: The management of superficial primary and recurrent esophageal cancer (EC) in medically inoperable patients is complex. Endoluminal high-dose-rate (HDR) brachytherapy has shown mixed results in terms of toxicity and local control. In this study, we examined the outcomes and toxicities in a set of patients with superficial primary and recurrent EC treated with a consistent HDR technique. METHODS AND MATERIALS: Between 8/2008 and 7/2011, 14 patients were treated with HDR intraluminal brachytherapy, 10 (71.4%) with recurrent disease, and 4 (28.6%) with previously unirradiated lesions. Patients received three weekly fractions to a median dose of 12 Gy (range, 10e15 Gy); dose was prescribed to 7-mm median depth with mucosal dose limited to 8e10 Gy using a 12e14-mm applicator. RESULTS: Median followup was 15.4 months. Overall freedom from failure (OFFF) and overall survival (OS) at 18 months were 30.8% (95% confidence interval [CI]: 5.2, 56.4) and 72.7% (95% CI: 45.3, 100), respectively. For patients with recurrent disease, OFFF and OS at 18 months were 11.1% (95% CI: 0, 32.1) and 55.6% (95% CI: 15.4, 95.8), respectively. For patients with previously unirradiated disease, OFFF and OS at 18 months were 75.0% (95% CI: 31.6, 100) and 100.0%, respectively. Eight (57.1%) patients had Grade 1 acute adverse effects; 6 (42.9%) patients had chronic Grade 1 adverse effects; 1 (7.1%) patient developed Grade 2 stricture. Grade 3 tracheoesophageal fistula occurred in 1 (7.1%) patient. One patient died before completion of treatment of unrelated causes. CONCLUSIONS: HDR endoluminal brachytherapy is a well-tolerated treatment for superficial primary and recurrent EC in medically inoperable patients. Ó 2013 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.
Keywords:
Endoluminal; HDR; Brachytherapy; Esophageal; Recurrent cancer; Medically inoperable
Introduction
Received 7 November 2012; received in revised form 21 December 2012; accepted 27 December 2012. This was presented as a poster presentation in part at the Annual European Society for Therapeutic Radiology and Oncology (ESTRO)/American Brachytherapy Society World Congress of Brachytherapy meeting in Barcelona, Spain on May 10e12, 2012. * Corresponding author. Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, Box 22, New York, NY 10065. Tel.: þ1-212-639-3983; fax: þ1-212-639-2417. E-mail address: [email protected] (K.A. Goodman).
Concurrent chemoradiation therapy is a component of standard management of locally advanced esophageal cancer (EC), both as definitive nonsurgical therapy (1e3) and as neoadjuvant treatment before surgery (4, 5). Local failure (LF) rates remain high, and the management of recurrent EC after external beam radiation therapy (EBRT) is complex. Patients with primary disease who are poor surgical candidates also represent a therapeutic challenge. Endoluminal high-dose-rate (HDR) brachytherapy has been implemented in these settings using a wide range of
1538-4721/$ - see front matter Ó 2013 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.brachy.2012.12.001
464
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
techniques and dosing regimens, with mixed results in terms of toxicity and local control. HDR endoluminal brachytherapy is a highly effective means of delivering tumoricidal doses to superficial lesions of the esophagus, while delivering much lower doses to surrounding tissues. In the setting of prior radiation, this localized delivery of dose is especially important as standard fields generally deliver significant dose to normal tissues, including the lungs and spinal cord. The spinal cord is generally treated to tolerance, leaving little room for additional radiation exposure. However, the rapid dose falloff seen in standard single-channel endoluminal treatments has disadvantages as well, as the dose delivered to the mucosal surface is much higher, often two- to threefold higher, than that delivered to the prescription depth. The American Brachytherapy Society (ABS) designed guidelines to assist in the use of HDR in the definitive and palliative treatment of ECs (6). In addition to selection criteria useful in determining which patients could potentially benefit from endoluminal esophageal brachytherapy, these guidelines specify the use of an applicator with an external diameter of 6e10 mm and dosing regimens between 5 and 10 Gy per fraction in one to four fractions based on clinical scenario. Using a prescription depth specified as 1 cm from the midsource or middwell without optimization, with a typical dose falloff, the mucosal surfaces can receive doses in excess of 15 Gy per fraction using the 6-mm applicator and assuming a 5-Gy fraction size. From the high-dose single-fraction external beam radiosurgery experience for spine metastases, it has recently been shown that doses in excess of 15 Gy result in a significantly greater risk of Grade 3 or greater toxicity (Fig. 1) (7). High rates of ulceration, stricture, and fistula have been observed at similar mucosal doses (8e10) using brachytherapy, which has curbed much of the interest in pursuing endoluminal brachytherapy as a treatment modality for EC. At the same time, these and other studies with lower
Fig. 1. Dose-response model for $Grade 3 esophagitis with singlefraction spine radiosurgery from logistic regression using the maximum dose to 2.5 cm3 of esophagus. Note the sharp increase in probability of $Grade 3 esophagitis at doses above 15 Gy. Reprinted from Cox et al. (7). Copyright 2012, with permission from Elsevier.
mucosal doses (11e13) have suggested that there may be a benefit in terms of local control and symptomatic palliation. As such, it seems that there may be a point where the therapeutic ratio is of benefit to patients, and efforts to determine the optimal means of delivering sufficient prescription dose while maintaining a safe mucosal dose are justified. In this study, we examined the outcomes and toxicities in a cohort of patients with early stage and locally recurrent EC treated with a consistent HDR technique designed to deliver adequate dose at prescription depth while limiting the mucosal dose to tolerable levels. In addition, we formulated a reference table (Table 1) of calculated mucosal doses based on applicator diameter and treatment depth using standard fraction sizes and treatment lengths to aid in a dosimetric comparison of mucosal doses delivered in the reported literature.
Methods Patients and procedure An intraluminal HDR brachytherapy procedure was developed to treat patients with recurrent EC after chemoradiation therapy with or without surgical resection or patients with superficial primary EC, who were not candidates for surgical resection and/or full-dose salvage chemotherapy. Between August 2008 and July 2011, 14 patients were treated with the described technique. After induction of general anesthesia with endotracheal intubation, the tumor extent is verified endoscopically; the proximal and distal extent of the tumor is then delineated fluoroscopically, and radiopaque fiducial markers are placed on the skin surface. A guidewire is inserted through the scope into the stomach, and the endoscope is then removed. A bougie catheter (generally 14-mm diameter; Varian Medical Systems, Inc., Palo Alto, CA), similar to a SavaryeGilliard (Cook Medical, Inc., Bloomington, IN) dilator, is passed over the guidewire, and the guidewire is then removed (Fig. 2). Dummy wires are placed in the central catheter and position verified; targeting of a longitudinal area with a 1-cm margin and offset from catheter tip is performed (treatment length is usually 4e8 cm). Dose is prescribed to median depth of 7 mm (range, 4e10 mm) from the surface of the bougie catheter, with prescription depth determined on the basis of clinical assessment of tumor thickness. A treatment plan is generated to provide the specified dose, generally limiting the mucosal surface dose to 8e10 Gy, and verified (Fig. 3). The treatment planning system used was ABACUS v3.1 (Varian, previously Isotopen-Technik Dr Sauerwein, Haan, Germany), and the plan is independently checked as previously described by Cohen et al. (14, 15). The patient is then connected to a GammaMed Plus remote afterloader (Varian), the room is cleared, and treatment is delivered. At the completion of treatment, the patient is then surveyed for residual activity, the catheter
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
465
Table 1 Calculated mucosal doses (assuming 5-cm treatment length, 5 Gy per fraction) Applicator outer diameter (mm)
Prescribed to 10 mm from source
Prescribed to 5 mm from applicator surface
Maximal dose Gy (% prescription)
Central dose Gy (% prescription)
Maximal dose Gy (% prescription)
Central dose Gy (% prescription)
4 5 6 8 10 12 14 15 16 20
78 53 38 23 16 12 9 8 7 5
32 23 19 13 11 9 7 7 6 5
43 31 26 23 16 13 12 11 11 10
23 19 15 14 11 10 9 9 9 8
(1550) (1054) (768) (454) (314) (232) (178) (158) (144) (104)
(636) (458) (372) (266) (216) (176) (148) (138) (128) (100)
and bougie are removed, and anesthesia is reversed. Treatments were delivered in three fractions over 3 weeks to a median dose of 12 Gy (range, 10e15 Gy). Chemotherapy For patients receiving concurrent chemotherapy, they received capecitabine at a dose of 1000 mg given orally twice daily 5 days per week during the duration of their treatment. Assessment/followup Patients were followed with physical examination, CTbased imaging, and upper gastrointestinal endoscopies every 2e3 months, with biopsies performed of any suspicious lesions observed. Grading of toxicity was based on the Common Terminology Criteria for Adverse Events, version 3.0, with the highest grade of any observed toxicity reported for each patient. Statistics All outcome measurements were measured from the completion of brachytherapy to the time of event. Overall
(856) (628) (524) (460) (314) (266) (232) (226) (218) (192)
(450) (372) (308) (272) (216) (200) (188) (185) (183) (162)
survival was defined as the time of death from any cause. LF is recurrence of disease at the site specifically treated with HDR intraluminal brachytherapy. Overall freedom from failure (OFFF) was defined as the time to any local, regional, or distant failure; freedom from local failure (FFLF) and freedom from distant metastasis (FFDM) were defined as the time to first clinical or pathologic evidence of LF or distant disease progression, respectively. Patients were censored at the date of the last followup or death. Outcomes were estimated using the KaplaneMeier statistics. Cox proportional hazards regression models were used for univariate and multivariate analyses. Statistical analysis was performed using SPSS, v.20.0.0 (SPSS, Inc., Chicago, IL). Dosimetric comparison To approximate the mucosal dose delivered in prior studies, the effective dose to the mucosa was simulated for a range of applicator diameters and treated depths. In these simulation plans, no attempt was made to restrict the mucosal dose as per our standard practice; instead, plans were designed to deliver a uniform dose at the prescription line. As expected, this created hot spots at
Fig. 2. Fluoroscopic films showing (a) guidewire placement and (b) insertion of dummy catheters.
466
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
capecitabine. Mean age at treatment was 76.4 years (Table 2). Outcomes Median followup for all patients was 15.4 months (range, 6.1e30.7 months); for all patients, FFLF at 18 months was 46.2% (95% confidence interval [CI]: 18.6, 73.8). FFDM and OFFF at 18 months were 83.3% (95% CI: 61.7, 100) and 30.8% (95% CI: 5.2, 56.4), respectively; OS at 12 and 18 months was 83.1% (95% CI: 61.1, 100) and 72.7% (95% CI: 45.3, 100), respectively (Fig. 4aed). Median time to LF was 3.4 months (range, 2.7e10 months), and median time to death was 13.5 months (range, 7.8e18.5 months). Table 3 shows the FFLF, FFDM, OFFF, and OS outcomes at 18 months for patients with recurrent disease and those treated in the upfront setting. Although there was no significant difference between outcomes for recurrent or primary Table 2 Patient characteristics Fig. 3. Two-dimensional isodose contours for a single high-dose-rate treatment; dose was prescribed to 7 mm, with mucosal surface limited to 9 Gy.
the distal and proximal ends of the treatment segment. The dose in the ends of the treated segments at the surface of the applicator will be referred to as the maximal dose; the dose at the geometric center of the treated segment at the surface of the applicator will be referred to as the central dose. Using ABACUS v3.1 (Varian) in the same manner as our standard treatment planning, and assuming a treated length of 5 cm and a prescription dose of 5 Gy, maximum and central doses were calculated; the applicator surface was conservatively considered a surrogate for or equivalent to the mucosal surface. Mucosal doses were tabulated for a range of applicator diameters and treated depths (either 5 mm from applicator surface or 10 mm from the center of the catheter).
Results Patient characteristics Between August 2008 and July 2011, 14 patients with EC were consented to treatment with HDR intraluminal brachytherapy, of whom 13 (93%) completed treatment and were further analyzed. Of the patients who completed treatment, 9 (69%) presented with recurrent disease after radiation and 4 (31%) presented with previously unirradiated lesions. Of the previously treated patients, median dose of prior radiation therapy was 50.4 Gy (range, 45e56 Gy). Histologies included 11 (85%) patients with adenocarcinoma and 2 (15%) with squamous cell carcinoma. Six (46%) patients also received concurrent chemotherapy with
Characteristics
N (%)
Total number of patients Mean age (range, y) Gender Male Female Prior RT Yes No Median dose of prior RT (range in Gy) Histology Squamous Adenocarcinoma Grade 1 2 3 Location of lesion Proximal/cervical Midthoracic Distal/GEJ Disease status Recurrent Primary Median dose of HDR (range in Gy) Median mucosal dose Concurrent chemotherapy Yes No Median followup (range in months) Status at last followup NED Locally recurrent, AWD Regional failure, AWD DOD DOC
14 76.4 (49.7e89) 8 (57.1) 6 (41.9) 10 (71.4) 4 (28.6) 50.4 (45e56) 2 (14.3) 12 (85.7) 3 (21.4) 9 (64.3) 2 (14.2) 1 (7.1) 3 (21.4) 10 (71.4) 10 4 12 10.4
(71.4) (28.6) (10e15) Gy
6 (42.9) 8 (57.1) 15.4 (6.1e30.7) 4 4 1 4 1
(28.6) (28.6) (7.1) (28.6) (28.6)
RT 5 radiation therapy; GEJ 5 gastroesophageal junction; HDR 5 high-dose-rate brachytherapy; NED 5 no evidence of disease; AWD 5 alive with disease; DOD 5 dead of disease; DOC 5 dead of other causes.
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
467
Fig. 4. Patient outcomes: (a) Rate of local recurrence, (b) freedom from distant metastasis, (c) overall freedom from failure, and (d) overall survival.
disease, there was a trend toward improved OFFF and OS in the primary disease group. Freedom from failure at 18 months for recurrent patients was 11.1% (95% CI: 0, 32.1) and primary patients was 75% (95% CI: 31.6, 100), logrank p 5 0.097; OS at 18 months for recurrent patients was 50% (95% CI: 12.4, 87.6) and for primary patients was 100%, log-rank p 5 0.051. Univariate analysis was performed to determine contribution of various risk factors on local recurrence, distant metastasis, and OS, including age older than 70 years, EC histology, primary vs. recurrent status, total brachytherapy dose delivered, use of concurrent chemotherapy, and prior radiation. None of these factors were found to be significant predictors of outcome.
developed Grade 2 chronic dysphagia. One of 14 (7.1%) patients developed an asymptomatic Grade 2 stricture. One of 14 (7.1%) patients developed a Grade 3 stricture requiring dilatation and subsequently developed a Grade 3 tracheoesophageal fistula; of note, this patient had received 50.4 Gy to the postoperative neck 14 years before his diagnosis of EC and then received 45 Gy to the upper esophagus before receiving HDR endoluminal brachytherapy. One patient died before completion of therapy because of conditions unrelated to her treatment. There was no significant difference in any acute or chronic toxicity for patients who received concurrent chemotherapy ( p 5 0.198 and 0.159, respectively) or in the incidence of Grade 2 or higher toxicity ( p 5 0.113).
Toxicity Eight of 14 (57.1%) patients had Grade 1 acute adverse effects (dysphagia, cough, or bleeding). There were no Grade 2 or higher acute adverse effects. Six of 14 (42.9%) patients had chronic Grade 1 adverse effects (dysphagia, cough, or bleeding); 1 of 14 (7.1%) patients
Mucosal dosimetric comparison In our patient cohort, the median mucosal dose (central dose) was 9.4 Gy. For most of the treated lesions treated superficially, the dose delivered during standard treatment was 9.2 Gy (range, 7.1e10.7 Gy). For 2 patients with more
Table 3 Outcomes at 18 months for treated patients Patient cohort
Freedom from local failure (%)
Freedom from distant metastasis (%)
Overall freedom from failure (%)
Overall survival (%)
Overall
46.2 95% CI: 18.6, 73.8 33.3 95% CI: 1.9, 64.7 75 95% CI: 31.6, 100
83.3 95% CI: 61.7, 100 75 95% CI: 44.4, 100 100
30.8 95% CI: 5.2, 56.4 11.1 95% CI: 0, 32.1 75 95% CI: 31.6, 100
72.7 95% CI: 45.3, 100 55.6 95% CI: 15.4, 95.8 100
Recurrent Primary
95% CI 5 95% confidence interval.
4.0 0.0 10.0 NR 0.0 12.2 5.0 6.7 1.4 NR 25.0 7.1 NR NR NR NR 7.6 NR NR NR NR NR 20.0 0.0 10 10 4e6* 10 6 10 16e20 13e15 10e20 (median 20) 15 4e6 10 6 10 4e6 10 10 10 20 10 6e8 10 14 12
Reference
Hishikawa et al. (8) Sur et al. (16) Sur et al. (17) Akagi et al. (11) Nishimura et al. (19) Gaspar et al. (9) Sharma et al. (10) Sur et al. (20) Vuong et al. (12) Nonoshita et al. (13) Rosenblatt et al. (18) Present study
NR 5 not reported. * Dimensions not given in reference; assumed based on standard Nucletron selectron (Nucletron, Netherlands) esophageal bougie size available at the time of the study.
10.0 8.0 28.7 NR NR NR 15.0 18.3 10.0 NR 10.0 7.1 28.0 32.0 NR NR 15.4 NR 10.0 NR 7 NR NR 0.0 3.4 0.0 12.8 0.0 0.0 6.1 NR NR 1.2 0.0 2.3 0 4.7 0.0 10.0 0.0 0.0 22.4 NR NR 1.4 0.0 14.2 7.1 12.96 22.3 22.3, 29.8, 22.3 3.24e9.15 6.48 18.6e31.8 22.3 29.8e50.9 8.64 4e5 21.3e29.8 9 NR NR NR NR 6.5 NR NR NR NR 4e5 16e27 9
Dose prescribed/ Reported fraction mucosal (Gy) dose (Gy) Prescription depth from surface of applicator (mm) Prescription depth from center of applicator (mm)
Table 4 Projected mucosal doses of historic studies and observed toxicity
Endoluminal brachytherapy may play a role in the management of locally recurrent or early stage EC in patients without surgical options. Our results demonstrate minimal toxicity with this three-fraction approach using applicators with diameters O12 mm and limiting the mucosal dose to !12 Gy. Although guidelines have been established both in terms of appropriate patient selection and appropriate technique (6), the use of HDR for EC in the practical clinical setting has been very heterogeneous, with applicators of widely varying dimensions, inconsistency in treated depth, and drastically different dosing regimens. According to ABS guidelines (6), good candidates for endoluminal HDR brachytherapy include those patients with a primary tumor #10 cm in length, confined to the esophageal wall, within the thoracic esophagus, and without regional lymph node or systemic metastases; patients without these characteristics were felt to be poor candidates, and in patients with esophageal fistulae, cervical esophagus involvement, or nontraversable stenoses, brachytherapy was considered to be contraindicated. At the time of their report, there was limited published experience in the use of brachytherapy in the cervical esophagus, and concern for development of a tracheoesophageal fistula prompted the contraindication. Table 4 summarizes the technique and toxicity associated with a number of relevant studies using endoluminal HDR brachytherapy, including the reported mucosal dose (if available) and what the calculated mucosal dose would have been based on our series of simulated geometries. Of note, several studies, including those by Hishikawa et al., Sur et al., Gaspar et al., and Rosenblatt et al. (8e10, 16e18), were performed with small-diameter applicators resulting in high mucosal doses (O12 Gy). These studies had very high toxicity and mixed outcomes, with ulceration rates of 28e32%, stricture rates of 8e25%, fistulae rates up to 25%, and treatment-related death in up to 13% of patients. Given these high rates of toxicity, it is understandable that practitioners would be reluctant to use HDR
Calculated mucosal dose/fraction (Gy), central
Discussion
6 6 6, 8, 6 2e5 4 5 6 8 4 4e5 8 5
bulky distal gastroesophageal junction lesions, prescription depth was set to 10 mm, resulting in higher mucosal doses on the order of 12.2e14 Gy. Table 1 summarizes the range of possible calculated mucosal doses (again, assuming a 5-cmetreated length and prescription dose of 5 Gy per fraction) based on variation of catheter diameter and prescription depth. Mucosal doses ranged from as low as 5 Gy (100% prescription dose) for a 20-mm diameter applicator with a prescription depth of 10 mm from the applicator surface (equivalent to prescribing to the mucosa) to as high as 31.8 Gy (636% prescription) for a 4-mm applicator with a prescription depth of 8 mm (10 mm from the middwell position).
5 7* 7 5 5 7e8 7 7e8 5 0 6e7 5
Local Life-threatening/ toxicity Stenosis serious toxicity resulting Ulceration Stricture (G2þ) Fistula (%) in death (%) (G2þ) (%) (G2þ) (%) (%) (%)
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
Applicator outer diameter (mm)
468
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
endoluminal brachytherapy more widely. However, there is evidence that different treatment geometries can achieve positive oncologic and palliative outcomes without excessive toxicity. For example, Sharma et al. (10) and Nonoshita et al. (13) have both reported on palliation and salvage using HDR for EC. In the study by Sharma et al. (10), using a 6-mm diameter applicator, toxicities included strictures in 15% of patients, ulceration in 10%, and tracheoesophageal fistula in 5% of patients. In contrast, Nonoshita et al. (13) used a double-balloon applicator with an outside diameter of 20 mm with no Grade 3 or higher toxicities observed. Reviewing other studies in which larger diameter applicators were used and mucosal dose was kept under 12 Gy, there are multiple examples of positive outcome with limited toxicity (11e13, 19) (Table 4). Akagi et al. (11) used larger diameter applicators as a boost to EBRT and achieved 5-year local control rates of 57% with 8.7% Grade 3 and 2.9% Grade 4 toxicities. Nishimura et al. (19) treated patients with superficial esophageal squamous cell carcinoma with EBRT and supplemental esophageal brachytherapy, with 3-year local control and OS of 85% and 100%, respectively. The only reported toxicity was transient esophageal ulcers; all toxicities were noted in patients treated with smaller applicators (10e15 mm in outer diameter). Vuong et al. (12) reported on their use of HDR esophageal brachytherapy as a boost to EBRT. The 2-year local recurrence rates were 25%, and only 15% of the 53 patients treated on the study reported Grade 2 esophagitis. In our study, we used a relatively large diameter applicator (~14 mm), allowing the delivery of relatively high dose to the prescription depth while limiting the mucosal surfaces to !12 Gy. Although outcomes are still poor in terms of local control, they appear to be more promising in the subset of patients with previously untreated disease. The only significant (Grade 3 or higher) toxicity noted secondary to treatment was in a patient who had previously been irradiated for a head and neck malignancy, receiving 50.4 Gy to the bilateral neck, followed by reirradiation of the upper esophagus to 45 Gy before his HDR endoluminal brachytherapy treatment in the cervical esophagus, a location that would have been contraindicated using the 1997 ABS guidelines (6). The reference table of calculated mucosal doses based on applicator diameter and treatment depth using standard fraction sizes and treatment lengths (Table 1) may be a useful resource for brachytherapy practitioners in their selection of applicator sizes and depth doses in esophageal and other intraluminal treatment applications. The correlation between high esophageal mucosal dose and toxicity mirrors to that seen in the singlefraction stereotactic radiosurgery experience in the spine, with a sharp increase in $Grade 3 toxicity at doses above 15 Gy (7), suggesting that applicators under 8 mm in diameter should not be used in the treatment of EC. Based on these results and an understanding of the role of mucosal dose and its contribution to toxicity, more
469
regular use of endoluminal brachytherapy seems warranted in the upfront setting for limited disease and as salvage therapy for previously treated malignancies. Dose escalation with larger diameter applicators may allow for improved therapeutic coverage of esophageal lesions without exceeding mucosal tolerance. Additionally, there was no detriment noted with concurrent chemotherapy, contrary to recommendations made in the ABS guidelines; patients likely may continue to receive systemic treatment without adversely affecting toxicity or outcomes. Given the high rate of local recurrence and minimal toxicity in these patients, we are planning a prospective dose escalation trial for superficial primary and recurrent EC in medically inoperable patients to determine the optimal brachytherapy regimen and improve local control. To assist in improving radiation planning and delivery, we plan to explore incorporation of MRI-based simulation techniques to improve soft tissue contrast for tumor delineation and means of delivering more conformal dose to target lesions.
Conclusion EC continues to have a poor prognosis despite years of effort in improving chemotherapy, surgery, and radiation therapy. Many patients present with advanced disease, and the location of the esophageal lesions and proximity to nearby critical structures limits the ability to deliver therapeutic dose. Moreover, patients who have received prior radiation have additional limitations on the delivery of additional radiation because of dose tolerance and surgically altered anatomy. HDR brachytherapy with or without concurrent chemotherapy is a safe and well-tolerated treatment for both previously untreated and recurrent lesions of the esophagus. Given the high rate of local recurrence and minimal toxicity in these patients, we are planning a prospective dose escalation trial for recurrent EC to determine the optimal brachytherapy regimen. References [1] Herskovic A, Martz K, al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593e1598. [2] al-Sarraf M, Martz K, Herskovic A, et al. Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: An intergroup study. J Clin Oncol 1997; 15:277e284. [3] Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer: Long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA 1999;281:1623e1627. [4] Tepper J, Krasna MJ, Niedzwiecki D, et al. Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol 2008;26:1086e1092. [5] Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of preoperative chemotherapy compared with chemoradiotherapy in patients
470
[6]
[7]
[8]
[9]
[10]
[11]
[12]
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470 with locally advanced adenocarcinoma of the esophagogastric junction. J Clin Oncol 2009;27:851e856. Gaspar LE, Nag S, Herskovic A, et al. American Brachytherapy Society (ABS) consensus guidelines for brachytherapy for esophageal cancer. Int J Radiat Oncol Biol Phys 1997;38: 127e132. Cox BW, Jackson A, Hunt M, et al. Esophageal toxicity from highdose, single-fraction paraspinal stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 2012;83:e661ee667. Hishikawa Y, Kurisu K, Taniguchi M, et al. High-dose-rate intraluminal brachytherapy for esophageal cancer: 10 years experience in Hyogo College of Medicine. Radiother Oncol 1991;21:107e114. Gaspar LE, Winter K, Kocha WI, et al. A phase I/II study of external beam radiation, brachytherapy, and concurrent chemotherapy for patients with localized carcinoma of the esophagus (Radiation Therapy Oncology Group Study 9207). Cancer 2000; 88:988e995. Sharma V, Mahantshetty U, Dinshaw KA, et al. Palliation of advanced/recurrent esophageal carcinoma with high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 2002;52:310e315. Akagi Y, Hirokawa Y, Kagemoto M, et al. Optimum fractionation for high-dose-rate endoesophageal brachytherapy following external irradiation of early stage esophageal cancer. Int J Radiat Oncol Biol Phys 1999;43:525e530. Vuong T, Szego P, David M, et al. The safety and usefulness of highdose-rate endoluminal brachytherapy as a boost in the treatment of patients with esophageal cancer with external beam radiation with or without chemotherapy. Int J Radiat Oncol Biol Phys 2005;63: 758e764.
[13] Nonoshita T, Sasaki T, Hirata H, et al. High-dose-rate brachytherapy for previously irradiated patients with recurrent esophageal cancer. Radiat Med 2007;25:373e377. [14] Cohen GN, Amols HI, Zaider M. An independent dose-to-point calculation program for the verification of high-dose-rate brachytherapy planning. Int J Radiat Oncol Biol Phys 2000;48:1251e1258. [15] Cohen GN, Zaider M. Intraoperative radiation therapy (IORT). In: Thomadsen BR, Rivard MJ, Butler WM, editors. Brachytherapy physics. 2nd ed. Madison, WI: Medical Physics Publishing; 2005. p. 511e520. [16] Sur RK, Singh DP, Sharma SC, et al. Radiation therapy of esophageal cancer: Role of high dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1992;22:1043e1046. [17] Sur RK, Donde B, Levin VC, et al. Fractionated high dose rate intraluminal brachytherapy in palliation of advanced esophageal cancer. Int J Radiat Oncol Biol Phys 1998;40:447e453. [18] Rosenblatt E, Jones G, Sur RK, et al. Adding external beam to intraluminal brachytherapy improves palliation in obstructive squamous cell oesophageal cancer: A prospective multi-centre randomized trial of the International Atomic Energy Agency. Radiother Oncol 2010; 97:488e494. [19] Nishimura Y, Okuno Y, Ono K, et al. External beam radiation therapy with or without high-dose-rate intraluminal brachytherapy for patients with superficial esophageal carcinoma. Cancer 1999;86: 220e228. [20] Sur R, Donde B, Falkson C, et al. Randomized prospective study comparing high-dose-rate intraluminal brachytherapy (HDRILBT) alone with HDRILBT and external beam radiotherapy in the palliation of advanced esophageal cancer. Brachytherapy 2004;3:191e195.
Endoluminal high-dose-rate brachytherapy for early stage and recurrent esophageal cancer in medically inoperable patients Michael R. Folkert1, Gil’ad N. Cohen2, Abraham J. Wu1, Hans Gerdes3, Mark A. Schattner3, Arnold J. Markowitz3, Emmy Ludwig3, David H. Ilson4, Manjit S. Bains5, Michael J. Zelefsky1, Karyn A. Goodman1,* 1
Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 2 Department of Medical Physics, Memorial Sloan-Kettering Cancer Center, New York, NY 3 Department of Gastroenterology, Memorial Sloan-Kettering Cancer Center, New York, NY 4 Department of Medical Oncology, Memorial Sloan-Kettering Cancer Center, New York, NY 5 Department of Thoracic Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY
ABSTRACT
PURPOSE: The management of superficial primary and recurrent esophageal cancer (EC) in medically inoperable patients is complex. Endoluminal high-dose-rate (HDR) brachytherapy has shown mixed results in terms of toxicity and local control. In this study, we examined the outcomes and toxicities in a set of patients with superficial primary and recurrent EC treated with a consistent HDR technique. METHODS AND MATERIALS: Between 8/2008 and 7/2011, 14 patients were treated with HDR intraluminal brachytherapy, 10 (71.4%) with recurrent disease, and 4 (28.6%) with previously unirradiated lesions. Patients received three weekly fractions to a median dose of 12 Gy (range, 10e15 Gy); dose was prescribed to 7-mm median depth with mucosal dose limited to 8e10 Gy using a 12e14-mm applicator. RESULTS: Median followup was 15.4 months. Overall freedom from failure (OFFF) and overall survival (OS) at 18 months were 30.8% (95% confidence interval [CI]: 5.2, 56.4) and 72.7% (95% CI: 45.3, 100), respectively. For patients with recurrent disease, OFFF and OS at 18 months were 11.1% (95% CI: 0, 32.1) and 55.6% (95% CI: 15.4, 95.8), respectively. For patients with previously unirradiated disease, OFFF and OS at 18 months were 75.0% (95% CI: 31.6, 100) and 100.0%, respectively. Eight (57.1%) patients had Grade 1 acute adverse effects; 6 (42.9%) patients had chronic Grade 1 adverse effects; 1 (7.1%) patient developed Grade 2 stricture. Grade 3 tracheoesophageal fistula occurred in 1 (7.1%) patient. One patient died before completion of treatment of unrelated causes. CONCLUSIONS: HDR endoluminal brachytherapy is a well-tolerated treatment for superficial primary and recurrent EC in medically inoperable patients. Ó 2013 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.
Keywords:
Endoluminal; HDR; Brachytherapy; Esophageal; Recurrent cancer; Medically inoperable
Introduction
Received 7 November 2012; received in revised form 21 December 2012; accepted 27 December 2012. This was presented as a poster presentation in part at the Annual European Society for Therapeutic Radiology and Oncology (ESTRO)/American Brachytherapy Society World Congress of Brachytherapy meeting in Barcelona, Spain on May 10e12, 2012. * Corresponding author. Department of Radiation Oncology, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, Box 22, New York, NY 10065. Tel.: þ1-212-639-3983; fax: þ1-212-639-2417. E-mail address: [email protected] (K.A. Goodman).
Concurrent chemoradiation therapy is a component of standard management of locally advanced esophageal cancer (EC), both as definitive nonsurgical therapy (1e3) and as neoadjuvant treatment before surgery (4, 5). Local failure (LF) rates remain high, and the management of recurrent EC after external beam radiation therapy (EBRT) is complex. Patients with primary disease who are poor surgical candidates also represent a therapeutic challenge. Endoluminal high-dose-rate (HDR) brachytherapy has been implemented in these settings using a wide range of
1538-4721/$ - see front matter Ó 2013 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.brachy.2012.12.001
464
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
techniques and dosing regimens, with mixed results in terms of toxicity and local control. HDR endoluminal brachytherapy is a highly effective means of delivering tumoricidal doses to superficial lesions of the esophagus, while delivering much lower doses to surrounding tissues. In the setting of prior radiation, this localized delivery of dose is especially important as standard fields generally deliver significant dose to normal tissues, including the lungs and spinal cord. The spinal cord is generally treated to tolerance, leaving little room for additional radiation exposure. However, the rapid dose falloff seen in standard single-channel endoluminal treatments has disadvantages as well, as the dose delivered to the mucosal surface is much higher, often two- to threefold higher, than that delivered to the prescription depth. The American Brachytherapy Society (ABS) designed guidelines to assist in the use of HDR in the definitive and palliative treatment of ECs (6). In addition to selection criteria useful in determining which patients could potentially benefit from endoluminal esophageal brachytherapy, these guidelines specify the use of an applicator with an external diameter of 6e10 mm and dosing regimens between 5 and 10 Gy per fraction in one to four fractions based on clinical scenario. Using a prescription depth specified as 1 cm from the midsource or middwell without optimization, with a typical dose falloff, the mucosal surfaces can receive doses in excess of 15 Gy per fraction using the 6-mm applicator and assuming a 5-Gy fraction size. From the high-dose single-fraction external beam radiosurgery experience for spine metastases, it has recently been shown that doses in excess of 15 Gy result in a significantly greater risk of Grade 3 or greater toxicity (Fig. 1) (7). High rates of ulceration, stricture, and fistula have been observed at similar mucosal doses (8e10) using brachytherapy, which has curbed much of the interest in pursuing endoluminal brachytherapy as a treatment modality for EC. At the same time, these and other studies with lower
Fig. 1. Dose-response model for $Grade 3 esophagitis with singlefraction spine radiosurgery from logistic regression using the maximum dose to 2.5 cm3 of esophagus. Note the sharp increase in probability of $Grade 3 esophagitis at doses above 15 Gy. Reprinted from Cox et al. (7). Copyright 2012, with permission from Elsevier.
mucosal doses (11e13) have suggested that there may be a benefit in terms of local control and symptomatic palliation. As such, it seems that there may be a point where the therapeutic ratio is of benefit to patients, and efforts to determine the optimal means of delivering sufficient prescription dose while maintaining a safe mucosal dose are justified. In this study, we examined the outcomes and toxicities in a cohort of patients with early stage and locally recurrent EC treated with a consistent HDR technique designed to deliver adequate dose at prescription depth while limiting the mucosal dose to tolerable levels. In addition, we formulated a reference table (Table 1) of calculated mucosal doses based on applicator diameter and treatment depth using standard fraction sizes and treatment lengths to aid in a dosimetric comparison of mucosal doses delivered in the reported literature.
Methods Patients and procedure An intraluminal HDR brachytherapy procedure was developed to treat patients with recurrent EC after chemoradiation therapy with or without surgical resection or patients with superficial primary EC, who were not candidates for surgical resection and/or full-dose salvage chemotherapy. Between August 2008 and July 2011, 14 patients were treated with the described technique. After induction of general anesthesia with endotracheal intubation, the tumor extent is verified endoscopically; the proximal and distal extent of the tumor is then delineated fluoroscopically, and radiopaque fiducial markers are placed on the skin surface. A guidewire is inserted through the scope into the stomach, and the endoscope is then removed. A bougie catheter (generally 14-mm diameter; Varian Medical Systems, Inc., Palo Alto, CA), similar to a SavaryeGilliard (Cook Medical, Inc., Bloomington, IN) dilator, is passed over the guidewire, and the guidewire is then removed (Fig. 2). Dummy wires are placed in the central catheter and position verified; targeting of a longitudinal area with a 1-cm margin and offset from catheter tip is performed (treatment length is usually 4e8 cm). Dose is prescribed to median depth of 7 mm (range, 4e10 mm) from the surface of the bougie catheter, with prescription depth determined on the basis of clinical assessment of tumor thickness. A treatment plan is generated to provide the specified dose, generally limiting the mucosal surface dose to 8e10 Gy, and verified (Fig. 3). The treatment planning system used was ABACUS v3.1 (Varian, previously Isotopen-Technik Dr Sauerwein, Haan, Germany), and the plan is independently checked as previously described by Cohen et al. (14, 15). The patient is then connected to a GammaMed Plus remote afterloader (Varian), the room is cleared, and treatment is delivered. At the completion of treatment, the patient is then surveyed for residual activity, the catheter
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
465
Table 1 Calculated mucosal doses (assuming 5-cm treatment length, 5 Gy per fraction) Applicator outer diameter (mm)
Prescribed to 10 mm from source
Prescribed to 5 mm from applicator surface
Maximal dose Gy (% prescription)
Central dose Gy (% prescription)
Maximal dose Gy (% prescription)
Central dose Gy (% prescription)
4 5 6 8 10 12 14 15 16 20
78 53 38 23 16 12 9 8 7 5
32 23 19 13 11 9 7 7 6 5
43 31 26 23 16 13 12 11 11 10
23 19 15 14 11 10 9 9 9 8
(1550) (1054) (768) (454) (314) (232) (178) (158) (144) (104)
(636) (458) (372) (266) (216) (176) (148) (138) (128) (100)
and bougie are removed, and anesthesia is reversed. Treatments were delivered in three fractions over 3 weeks to a median dose of 12 Gy (range, 10e15 Gy). Chemotherapy For patients receiving concurrent chemotherapy, they received capecitabine at a dose of 1000 mg given orally twice daily 5 days per week during the duration of their treatment. Assessment/followup Patients were followed with physical examination, CTbased imaging, and upper gastrointestinal endoscopies every 2e3 months, with biopsies performed of any suspicious lesions observed. Grading of toxicity was based on the Common Terminology Criteria for Adverse Events, version 3.0, with the highest grade of any observed toxicity reported for each patient. Statistics All outcome measurements were measured from the completion of brachytherapy to the time of event. Overall
(856) (628) (524) (460) (314) (266) (232) (226) (218) (192)
(450) (372) (308) (272) (216) (200) (188) (185) (183) (162)
survival was defined as the time of death from any cause. LF is recurrence of disease at the site specifically treated with HDR intraluminal brachytherapy. Overall freedom from failure (OFFF) was defined as the time to any local, regional, or distant failure; freedom from local failure (FFLF) and freedom from distant metastasis (FFDM) were defined as the time to first clinical or pathologic evidence of LF or distant disease progression, respectively. Patients were censored at the date of the last followup or death. Outcomes were estimated using the KaplaneMeier statistics. Cox proportional hazards regression models were used for univariate and multivariate analyses. Statistical analysis was performed using SPSS, v.20.0.0 (SPSS, Inc., Chicago, IL). Dosimetric comparison To approximate the mucosal dose delivered in prior studies, the effective dose to the mucosa was simulated for a range of applicator diameters and treated depths. In these simulation plans, no attempt was made to restrict the mucosal dose as per our standard practice; instead, plans were designed to deliver a uniform dose at the prescription line. As expected, this created hot spots at
Fig. 2. Fluoroscopic films showing (a) guidewire placement and (b) insertion of dummy catheters.
466
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
capecitabine. Mean age at treatment was 76.4 years (Table 2). Outcomes Median followup for all patients was 15.4 months (range, 6.1e30.7 months); for all patients, FFLF at 18 months was 46.2% (95% confidence interval [CI]: 18.6, 73.8). FFDM and OFFF at 18 months were 83.3% (95% CI: 61.7, 100) and 30.8% (95% CI: 5.2, 56.4), respectively; OS at 12 and 18 months was 83.1% (95% CI: 61.1, 100) and 72.7% (95% CI: 45.3, 100), respectively (Fig. 4aed). Median time to LF was 3.4 months (range, 2.7e10 months), and median time to death was 13.5 months (range, 7.8e18.5 months). Table 3 shows the FFLF, FFDM, OFFF, and OS outcomes at 18 months for patients with recurrent disease and those treated in the upfront setting. Although there was no significant difference between outcomes for recurrent or primary Table 2 Patient characteristics Fig. 3. Two-dimensional isodose contours for a single high-dose-rate treatment; dose was prescribed to 7 mm, with mucosal surface limited to 9 Gy.
the distal and proximal ends of the treatment segment. The dose in the ends of the treated segments at the surface of the applicator will be referred to as the maximal dose; the dose at the geometric center of the treated segment at the surface of the applicator will be referred to as the central dose. Using ABACUS v3.1 (Varian) in the same manner as our standard treatment planning, and assuming a treated length of 5 cm and a prescription dose of 5 Gy, maximum and central doses were calculated; the applicator surface was conservatively considered a surrogate for or equivalent to the mucosal surface. Mucosal doses were tabulated for a range of applicator diameters and treated depths (either 5 mm from applicator surface or 10 mm from the center of the catheter).
Results Patient characteristics Between August 2008 and July 2011, 14 patients with EC were consented to treatment with HDR intraluminal brachytherapy, of whom 13 (93%) completed treatment and were further analyzed. Of the patients who completed treatment, 9 (69%) presented with recurrent disease after radiation and 4 (31%) presented with previously unirradiated lesions. Of the previously treated patients, median dose of prior radiation therapy was 50.4 Gy (range, 45e56 Gy). Histologies included 11 (85%) patients with adenocarcinoma and 2 (15%) with squamous cell carcinoma. Six (46%) patients also received concurrent chemotherapy with
Characteristics
N (%)
Total number of patients Mean age (range, y) Gender Male Female Prior RT Yes No Median dose of prior RT (range in Gy) Histology Squamous Adenocarcinoma Grade 1 2 3 Location of lesion Proximal/cervical Midthoracic Distal/GEJ Disease status Recurrent Primary Median dose of HDR (range in Gy) Median mucosal dose Concurrent chemotherapy Yes No Median followup (range in months) Status at last followup NED Locally recurrent, AWD Regional failure, AWD DOD DOC
14 76.4 (49.7e89) 8 (57.1) 6 (41.9) 10 (71.4) 4 (28.6) 50.4 (45e56) 2 (14.3) 12 (85.7) 3 (21.4) 9 (64.3) 2 (14.2) 1 (7.1) 3 (21.4) 10 (71.4) 10 4 12 10.4
(71.4) (28.6) (10e15) Gy
6 (42.9) 8 (57.1) 15.4 (6.1e30.7) 4 4 1 4 1
(28.6) (28.6) (7.1) (28.6) (28.6)
RT 5 radiation therapy; GEJ 5 gastroesophageal junction; HDR 5 high-dose-rate brachytherapy; NED 5 no evidence of disease; AWD 5 alive with disease; DOD 5 dead of disease; DOC 5 dead of other causes.
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
467
Fig. 4. Patient outcomes: (a) Rate of local recurrence, (b) freedom from distant metastasis, (c) overall freedom from failure, and (d) overall survival.
disease, there was a trend toward improved OFFF and OS in the primary disease group. Freedom from failure at 18 months for recurrent patients was 11.1% (95% CI: 0, 32.1) and primary patients was 75% (95% CI: 31.6, 100), logrank p 5 0.097; OS at 18 months for recurrent patients was 50% (95% CI: 12.4, 87.6) and for primary patients was 100%, log-rank p 5 0.051. Univariate analysis was performed to determine contribution of various risk factors on local recurrence, distant metastasis, and OS, including age older than 70 years, EC histology, primary vs. recurrent status, total brachytherapy dose delivered, use of concurrent chemotherapy, and prior radiation. None of these factors were found to be significant predictors of outcome.
developed Grade 2 chronic dysphagia. One of 14 (7.1%) patients developed an asymptomatic Grade 2 stricture. One of 14 (7.1%) patients developed a Grade 3 stricture requiring dilatation and subsequently developed a Grade 3 tracheoesophageal fistula; of note, this patient had received 50.4 Gy to the postoperative neck 14 years before his diagnosis of EC and then received 45 Gy to the upper esophagus before receiving HDR endoluminal brachytherapy. One patient died before completion of therapy because of conditions unrelated to her treatment. There was no significant difference in any acute or chronic toxicity for patients who received concurrent chemotherapy ( p 5 0.198 and 0.159, respectively) or in the incidence of Grade 2 or higher toxicity ( p 5 0.113).
Toxicity Eight of 14 (57.1%) patients had Grade 1 acute adverse effects (dysphagia, cough, or bleeding). There were no Grade 2 or higher acute adverse effects. Six of 14 (42.9%) patients had chronic Grade 1 adverse effects (dysphagia, cough, or bleeding); 1 of 14 (7.1%) patients
Mucosal dosimetric comparison In our patient cohort, the median mucosal dose (central dose) was 9.4 Gy. For most of the treated lesions treated superficially, the dose delivered during standard treatment was 9.2 Gy (range, 7.1e10.7 Gy). For 2 patients with more
Table 3 Outcomes at 18 months for treated patients Patient cohort
Freedom from local failure (%)
Freedom from distant metastasis (%)
Overall freedom from failure (%)
Overall survival (%)
Overall
46.2 95% CI: 18.6, 73.8 33.3 95% CI: 1.9, 64.7 75 95% CI: 31.6, 100
83.3 95% CI: 61.7, 100 75 95% CI: 44.4, 100 100
30.8 95% CI: 5.2, 56.4 11.1 95% CI: 0, 32.1 75 95% CI: 31.6, 100
72.7 95% CI: 45.3, 100 55.6 95% CI: 15.4, 95.8 100
Recurrent Primary
95% CI 5 95% confidence interval.
4.0 0.0 10.0 NR 0.0 12.2 5.0 6.7 1.4 NR 25.0 7.1 NR NR NR NR 7.6 NR NR NR NR NR 20.0 0.0 10 10 4e6* 10 6 10 16e20 13e15 10e20 (median 20) 15 4e6 10 6 10 4e6 10 10 10 20 10 6e8 10 14 12
Reference
Hishikawa et al. (8) Sur et al. (16) Sur et al. (17) Akagi et al. (11) Nishimura et al. (19) Gaspar et al. (9) Sharma et al. (10) Sur et al. (20) Vuong et al. (12) Nonoshita et al. (13) Rosenblatt et al. (18) Present study
NR 5 not reported. * Dimensions not given in reference; assumed based on standard Nucletron selectron (Nucletron, Netherlands) esophageal bougie size available at the time of the study.
10.0 8.0 28.7 NR NR NR 15.0 18.3 10.0 NR 10.0 7.1 28.0 32.0 NR NR 15.4 NR 10.0 NR 7 NR NR 0.0 3.4 0.0 12.8 0.0 0.0 6.1 NR NR 1.2 0.0 2.3 0 4.7 0.0 10.0 0.0 0.0 22.4 NR NR 1.4 0.0 14.2 7.1 12.96 22.3 22.3, 29.8, 22.3 3.24e9.15 6.48 18.6e31.8 22.3 29.8e50.9 8.64 4e5 21.3e29.8 9 NR NR NR NR 6.5 NR NR NR NR 4e5 16e27 9
Dose prescribed/ Reported fraction mucosal (Gy) dose (Gy) Prescription depth from surface of applicator (mm) Prescription depth from center of applicator (mm)
Table 4 Projected mucosal doses of historic studies and observed toxicity
Endoluminal brachytherapy may play a role in the management of locally recurrent or early stage EC in patients without surgical options. Our results demonstrate minimal toxicity with this three-fraction approach using applicators with diameters O12 mm and limiting the mucosal dose to !12 Gy. Although guidelines have been established both in terms of appropriate patient selection and appropriate technique (6), the use of HDR for EC in the practical clinical setting has been very heterogeneous, with applicators of widely varying dimensions, inconsistency in treated depth, and drastically different dosing regimens. According to ABS guidelines (6), good candidates for endoluminal HDR brachytherapy include those patients with a primary tumor #10 cm in length, confined to the esophageal wall, within the thoracic esophagus, and without regional lymph node or systemic metastases; patients without these characteristics were felt to be poor candidates, and in patients with esophageal fistulae, cervical esophagus involvement, or nontraversable stenoses, brachytherapy was considered to be contraindicated. At the time of their report, there was limited published experience in the use of brachytherapy in the cervical esophagus, and concern for development of a tracheoesophageal fistula prompted the contraindication. Table 4 summarizes the technique and toxicity associated with a number of relevant studies using endoluminal HDR brachytherapy, including the reported mucosal dose (if available) and what the calculated mucosal dose would have been based on our series of simulated geometries. Of note, several studies, including those by Hishikawa et al., Sur et al., Gaspar et al., and Rosenblatt et al. (8e10, 16e18), were performed with small-diameter applicators resulting in high mucosal doses (O12 Gy). These studies had very high toxicity and mixed outcomes, with ulceration rates of 28e32%, stricture rates of 8e25%, fistulae rates up to 25%, and treatment-related death in up to 13% of patients. Given these high rates of toxicity, it is understandable that practitioners would be reluctant to use HDR
Calculated mucosal dose/fraction (Gy), central
Discussion
6 6 6, 8, 6 2e5 4 5 6 8 4 4e5 8 5
bulky distal gastroesophageal junction lesions, prescription depth was set to 10 mm, resulting in higher mucosal doses on the order of 12.2e14 Gy. Table 1 summarizes the range of possible calculated mucosal doses (again, assuming a 5-cmetreated length and prescription dose of 5 Gy per fraction) based on variation of catheter diameter and prescription depth. Mucosal doses ranged from as low as 5 Gy (100% prescription dose) for a 20-mm diameter applicator with a prescription depth of 10 mm from the applicator surface (equivalent to prescribing to the mucosa) to as high as 31.8 Gy (636% prescription) for a 4-mm applicator with a prescription depth of 8 mm (10 mm from the middwell position).
5 7* 7 5 5 7e8 7 7e8 5 0 6e7 5
Local Life-threatening/ toxicity Stenosis serious toxicity resulting Ulceration Stricture (G2þ) Fistula (%) in death (%) (G2þ) (%) (G2þ) (%) (%) (%)
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
Applicator outer diameter (mm)
468
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470
endoluminal brachytherapy more widely. However, there is evidence that different treatment geometries can achieve positive oncologic and palliative outcomes without excessive toxicity. For example, Sharma et al. (10) and Nonoshita et al. (13) have both reported on palliation and salvage using HDR for EC. In the study by Sharma et al. (10), using a 6-mm diameter applicator, toxicities included strictures in 15% of patients, ulceration in 10%, and tracheoesophageal fistula in 5% of patients. In contrast, Nonoshita et al. (13) used a double-balloon applicator with an outside diameter of 20 mm with no Grade 3 or higher toxicities observed. Reviewing other studies in which larger diameter applicators were used and mucosal dose was kept under 12 Gy, there are multiple examples of positive outcome with limited toxicity (11e13, 19) (Table 4). Akagi et al. (11) used larger diameter applicators as a boost to EBRT and achieved 5-year local control rates of 57% with 8.7% Grade 3 and 2.9% Grade 4 toxicities. Nishimura et al. (19) treated patients with superficial esophageal squamous cell carcinoma with EBRT and supplemental esophageal brachytherapy, with 3-year local control and OS of 85% and 100%, respectively. The only reported toxicity was transient esophageal ulcers; all toxicities were noted in patients treated with smaller applicators (10e15 mm in outer diameter). Vuong et al. (12) reported on their use of HDR esophageal brachytherapy as a boost to EBRT. The 2-year local recurrence rates were 25%, and only 15% of the 53 patients treated on the study reported Grade 2 esophagitis. In our study, we used a relatively large diameter applicator (~14 mm), allowing the delivery of relatively high dose to the prescription depth while limiting the mucosal surfaces to !12 Gy. Although outcomes are still poor in terms of local control, they appear to be more promising in the subset of patients with previously untreated disease. The only significant (Grade 3 or higher) toxicity noted secondary to treatment was in a patient who had previously been irradiated for a head and neck malignancy, receiving 50.4 Gy to the bilateral neck, followed by reirradiation of the upper esophagus to 45 Gy before his HDR endoluminal brachytherapy treatment in the cervical esophagus, a location that would have been contraindicated using the 1997 ABS guidelines (6). The reference table of calculated mucosal doses based on applicator diameter and treatment depth using standard fraction sizes and treatment lengths (Table 1) may be a useful resource for brachytherapy practitioners in their selection of applicator sizes and depth doses in esophageal and other intraluminal treatment applications. The correlation between high esophageal mucosal dose and toxicity mirrors to that seen in the singlefraction stereotactic radiosurgery experience in the spine, with a sharp increase in $Grade 3 toxicity at doses above 15 Gy (7), suggesting that applicators under 8 mm in diameter should not be used in the treatment of EC. Based on these results and an understanding of the role of mucosal dose and its contribution to toxicity, more
469
regular use of endoluminal brachytherapy seems warranted in the upfront setting for limited disease and as salvage therapy for previously treated malignancies. Dose escalation with larger diameter applicators may allow for improved therapeutic coverage of esophageal lesions without exceeding mucosal tolerance. Additionally, there was no detriment noted with concurrent chemotherapy, contrary to recommendations made in the ABS guidelines; patients likely may continue to receive systemic treatment without adversely affecting toxicity or outcomes. Given the high rate of local recurrence and minimal toxicity in these patients, we are planning a prospective dose escalation trial for superficial primary and recurrent EC in medically inoperable patients to determine the optimal brachytherapy regimen and improve local control. To assist in improving radiation planning and delivery, we plan to explore incorporation of MRI-based simulation techniques to improve soft tissue contrast for tumor delineation and means of delivering more conformal dose to target lesions.
Conclusion EC continues to have a poor prognosis despite years of effort in improving chemotherapy, surgery, and radiation therapy. Many patients present with advanced disease, and the location of the esophageal lesions and proximity to nearby critical structures limits the ability to deliver therapeutic dose. Moreover, patients who have received prior radiation have additional limitations on the delivery of additional radiation because of dose tolerance and surgically altered anatomy. HDR brachytherapy with or without concurrent chemotherapy is a safe and well-tolerated treatment for both previously untreated and recurrent lesions of the esophagus. Given the high rate of local recurrence and minimal toxicity in these patients, we are planning a prospective dose escalation trial for recurrent EC to determine the optimal brachytherapy regimen. References [1] Herskovic A, Martz K, al-Sarraf M, et al. Combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. N Engl J Med 1992;326:1593e1598. [2] al-Sarraf M, Martz K, Herskovic A, et al. Progress report of combined chemoradiotherapy versus radiotherapy alone in patients with esophageal cancer: An intergroup study. J Clin Oncol 1997; 15:277e284. [3] Cooper JS, Guo MD, Herskovic A, et al. Chemoradiotherapy of locally advanced esophageal cancer: Long-term follow-up of a prospective randomized trial (RTOG 85-01). Radiation Therapy Oncology Group. JAMA 1999;281:1623e1627. [4] Tepper J, Krasna MJ, Niedzwiecki D, et al. Phase III trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: CALGB 9781. J Clin Oncol 2008;26:1086e1092. [5] Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of preoperative chemotherapy compared with chemoradiotherapy in patients
470
[6]
[7]
[8]
[9]
[10]
[11]
[12]
M.R. Folkert et al. / Brachytherapy 12 (2013) 463e470 with locally advanced adenocarcinoma of the esophagogastric junction. J Clin Oncol 2009;27:851e856. Gaspar LE, Nag S, Herskovic A, et al. American Brachytherapy Society (ABS) consensus guidelines for brachytherapy for esophageal cancer. Int J Radiat Oncol Biol Phys 1997;38: 127e132. Cox BW, Jackson A, Hunt M, et al. Esophageal toxicity from highdose, single-fraction paraspinal stereotactic radiosurgery. Int J Radiat Oncol Biol Phys 2012;83:e661ee667. Hishikawa Y, Kurisu K, Taniguchi M, et al. High-dose-rate intraluminal brachytherapy for esophageal cancer: 10 years experience in Hyogo College of Medicine. Radiother Oncol 1991;21:107e114. Gaspar LE, Winter K, Kocha WI, et al. A phase I/II study of external beam radiation, brachytherapy, and concurrent chemotherapy for patients with localized carcinoma of the esophagus (Radiation Therapy Oncology Group Study 9207). Cancer 2000; 88:988e995. Sharma V, Mahantshetty U, Dinshaw KA, et al. Palliation of advanced/recurrent esophageal carcinoma with high-dose-rate brachytherapy. Int J Radiat Oncol Biol Phys 2002;52:310e315. Akagi Y, Hirokawa Y, Kagemoto M, et al. Optimum fractionation for high-dose-rate endoesophageal brachytherapy following external irradiation of early stage esophageal cancer. Int J Radiat Oncol Biol Phys 1999;43:525e530. Vuong T, Szego P, David M, et al. The safety and usefulness of highdose-rate endoluminal brachytherapy as a boost in the treatment of patients with esophageal cancer with external beam radiation with or without chemotherapy. Int J Radiat Oncol Biol Phys 2005;63: 758e764.
[13] Nonoshita T, Sasaki T, Hirata H, et al. High-dose-rate brachytherapy for previously irradiated patients with recurrent esophageal cancer. Radiat Med 2007;25:373e377. [14] Cohen GN, Amols HI, Zaider M. An independent dose-to-point calculation program for the verification of high-dose-rate brachytherapy planning. Int J Radiat Oncol Biol Phys 2000;48:1251e1258. [15] Cohen GN, Zaider M. Intraoperative radiation therapy (IORT). In: Thomadsen BR, Rivard MJ, Butler WM, editors. Brachytherapy physics. 2nd ed. Madison, WI: Medical Physics Publishing; 2005. p. 511e520. [16] Sur RK, Singh DP, Sharma SC, et al. Radiation therapy of esophageal cancer: Role of high dose rate brachytherapy. Int J Radiat Oncol Biol Phys 1992;22:1043e1046. [17] Sur RK, Donde B, Levin VC, et al. Fractionated high dose rate intraluminal brachytherapy in palliation of advanced esophageal cancer. Int J Radiat Oncol Biol Phys 1998;40:447e453. [18] Rosenblatt E, Jones G, Sur RK, et al. Adding external beam to intraluminal brachytherapy improves palliation in obstructive squamous cell oesophageal cancer: A prospective multi-centre randomized trial of the International Atomic Energy Agency. Radiother Oncol 2010; 97:488e494. [19] Nishimura Y, Okuno Y, Ono K, et al. External beam radiation therapy with or without high-dose-rate intraluminal brachytherapy for patients with superficial esophageal carcinoma. Cancer 1999;86: 220e228. [20] Sur R, Donde B, Falkson C, et al. Randomized prospective study comparing high-dose-rate intraluminal brachytherapy (HDRILBT) alone with HDRILBT and external beam radiotherapy in the palliation of advanced esophageal cancer. Brachytherapy 2004;3:191e195.