Available online at www.sciencedirect.com
American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379 – 384 www.elsevier.com/locate/amjoto
Original contributions
Transoral robotic surgery for supraglottic squamous cell carcinoma☆,☆☆,★ Steven M. Olsen, MD, Eric J. Moore, MD⁎, Cody A. Koch, MD, PhD, Daniel L. Price, MD, Jan L. Kasperbauer, MD, Kerry D. Olsen, MD Department of Otorhinolaryngology–Head and Neck Surgery, Mayo Clinic, Rochester, MN, USA Received 24 August 2011
Abstract
Purpose: We present our experience with the use of transoral robotic surgery (TORS) for treatment of supraglottic squamous cell carcinoma. Materials and Methods: We studied all patients who underwent TORS for supraglottic squamous cell carcinoma, with or without adjuvant therapy, from March 2007 through June 2009, who had a minimum of 2 years of follow-up. Primary functional outcomes included dysphonia, tracheostomy dependence, and gastrostomy tube dependence. Disease control and survival were estimated with the Kaplan-Meier method. Results: Of 9 patients in the study group, 7 (78%) had advanced-stage disease. All 9 patients had negative margins after TORS, with no perioperative complications. Regional recurrence and local recurrence developed in 1 patient each. One patient died of disease. At last follow-up, 7 patients (78%) were tracheostomy free, and 7 (78%) were gastrostomy tube free. Conclusions: Transoral robotic surgery is a promising modality for resection of supraglottic squamous cell carcinoma. Transoral robotic surgery achieved functional laryngeal preservation in most patients with no complications. © 2012 Elsevier Inc. All rights reserved.
1. Introduction Treatment options for supraglottic squamous cell carcinoma (SG SCC) include operative therapy (open surgical procedures and transoral procedures), radiotherapy, chemoradiotherapy, or combinations of these. The optimal treatment is individualized to the patient, tumor subsite, and stage. The main goals of treatment include optimizing disease control and survival while minimizing functional morbidity [1]. Primary surgical options for SG SCC range from minimally invasive transoral approaches to total laryngectomy. Most SG SCCs that do not extend onto the vocal cords or cause cord ☆
Mayo Clinic does not endorse the products mentioned in this article. Conflict of Interest: None of the authors has any conflict of interest, financial or otherwise. ★ Presented at the 114th Annual Meeting of the Combined Otolaryngology Spring Meetings (COSM), Chicago, Illinois, April 29, 2011. ⁎ Corresponding author. Department of Otorhinolaryngology-Head and Neck Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905, USA. E-mail address:
[email protected] (E.J. Moore). ☆☆
0196-0709/$ – see front matter © 2012 Elsevier Inc. All rights reserved. doi:10.1016/j.amjoto.2011.10.007
fixation can be successfully treated with transoral or open partial laryngectomy. Oncologic outcomes for laryngeal cancer are equivalent for transoral approaches and open approaches [2]. The surgical approach is selected based on surgeon skill and training; patient factors including body habitus, pulmonary function, and oral and neck compliance; and tumor factors including location and extent. Surgeons are increasingly performing minimally invasive transoral laser microsurgery in an attempt to preserve laryngeal function without sacrificing disease control and survival. In appropriately selected patients, transoral laser microsurgery achieves oncologic outcomes comparable with those with open surgical approaches, with shorter hospitalizations, decreased morbidity, and improved functional outcomes [1,3]. A promising new modality to improve on the limitations of transoral laser microsurgery is transoral robotic surgery (TORS). We have used TORS for the treatment of select head and neck cancers since March 2007. The advantages of using TORS for resection of SG SCC include improved visualization, improved manual dexterity, tremor reduction, and the
380
S.M. Olsen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379–384
availability of many types of instruments including laser and cautery options and angulated instruments and endoscopes capable of working around corners. Several small case series have demonstrated that TORS is a feasible modality for resection of cancers of the larynx, but oncologic and functional outcomes are lacking [4-7]. Because TORS represents a new technique in head and neck cancer treatment, careful analysis of treatment-related morbidity, functional results, and oncologic outcomes is essential to ensure that results are comparable or superior to current widely accepted treatment options. Here, we describe disease control, survival, and functional outcomes in patients treated with TORS for SG SCC to justify continuation of this new treatment modality. 2. Materials and methods 2.1. Study design This study was approved by and conducted in compliance with the Mayo Clinic Institutional Review Board. All patients who underwent TORS as primary treatment for SG SCC from March 2007 through June 2009 at Mayo Clinic, Rochester, Minnesota, were included in this series. The time frame chosen ensured that patients were followed up for a minimum of 24 months after surgery. Data on patient demographics, tumor stage, histopathologic characteristics, treatment modalities, and functional and oncologic outcomes were collected from a prospectively compiled database. 2.2. Patient candidacy for TORS Patients were considered for TORS if they had a supraglottic tumor estimated to be resectable via the transoral route based on preoperative imaging and physical examination. Patients with retrognathia, narrow mandibular arch, macroglossia, or limited neck extension were considered poor candidates for TORS because of the difficulty of exposing the tumor and were offered alternative operative or nonoperative treatment. Patients with tumors that caused vocal cord fixation or those that extended transglottically, into the deep tongue musculature, hyoid, or lateral neck soft tissue were not considered candidates for TORS. 2.3. Surgical technique After securing the airway by standard endotracheal intubation with a laser-safe tube or tracheostomy placement, the patient was positioned for TORS, and the teeth and alveolar ridge were protected with a thermoplastic mouthguard. The larynx was exposed using a FeyhKastenbauer laryngeal retractor (Gyrus ACMI, Southborough, MA). Instrumentation used included a 30° binocular endoscope, a 5-mm EndoWrist Schertel grasper, and a 5-mm spatula cautery (Intuitive Surgical, Inc, Sunnyvale, CA). A surgical assistant was positioned at
the patient's head, aiding with suction, application of vascular clips (superior laryngeal vessels), and surgical specimen retrieval and pathologic orientation. Tumors were resected en block when possible. After removing the tumor, peripheral margins were examined in the frozen section pathology laboratory, and positive margins were reexcised until clear margins were achieved. After margins were cleared, the surgical site was checked for hemostasis, and a nasogastric tube was placed. Neck dissections were performed in all cases at the time of TORS. 2.4. Postoperative care Patients had their tracheostomy tube removed when they could tolerate corking for 48 continuous hours. All patients underwent repeat examination and speech-language pathology evaluation with a video swallow study within 30 days of surgery. If swallowing function was deemed adequate, the nasogastric tube was removed, and the patient was allowed to resume an oral diet. Patients with persistent dysphagia were considered for additional swallowing therapy or gastrostomy tube (G-tube) placement, depending on dysphagia severity and adjuvant therapy requirements. Functional speech outcomes were determined from the patient's most recent speech-language pathology evaluation. Speech quality was recorded for 3 general parameters: phonation, resonance, and articulation. Only phonation was affected in this series. Phonation characteristics including hoarseness, strain, pitch, wetness, and breathiness were scored on a 0 (normal) to 4 (profoundly impaired) scale by a speechlanguage pathologist. Scores from each category were then added, and the patient's speech was categorized as follows: 0, normal speech; 1 to 2, mild dysphonia; 3 to 5, moderate dysphonia; or 6 or greater, severe dysphonia. 2.5. Adjuvant therapy Adjuvant therapy included intensity-modulated radiotherapy with or without chemotherapy. Indications for adjuvant therapy were usually determined by neck disease and included extracapsular spread, desmoplasia, multiple positive lymph nodes, or atypical metastatic patterns. Indications for adjuvant therapy based on the primary tumor included lymphovascular invasion, perineural invasion, grade 4/4 disease, and positive resection margins. The decision to pursue adjuvant therapy was based on the final pathology report and consultation with medical and radiation oncologists. 2.6. Statistics Data are summarized as number of patients (percentage). Commercially available statistical software was used (JMP; SAS Institute, Inc, Cary, NC). Oncologic outcomes were estimated using Kaplan-Meier regression analysis. Patients were censored on the date of death or the date of last contact.
S.M. Olsen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379–384
381
Table 2 AJCC clinical and pathologic staging
3. Results 3.1. Patients
Stage
During the study period, 9 patients were treated with TORS for biopsy-proven SG SCCA (Table 1). Seven patients (78%) were men, and the mean age was 61.9 years (range, 45–77 years). Seven patients (78%) had a history of tobacco, use and 4 patients (44%) had a history of daily alcohol use. 3.2. Tumors Tumors were staged in accordance with American Joint Committee on Cancer guidelines (Table 2). Preoperative clinical staging and pathologic staging varied among the group. Four patients had negative lymph nodes, and 5 patients had lymph node metastases. On overall staging, most patients (78%) had advanced disease (stage III-IVa). Histopathologic review of the primary tumors demonstrated angiolymphatic invasion in 1 patient and perineural invasion in 1 patient. Histopathologic analysis of lymph node metastases demonstrated desmoplasia in 3 patients and extracapsular spread in 3 patients.
Clinical T stage T1 T2 T3 T4a Pathologic T stage T1 T2 T3 T4a Pathologic N stage N0 N1 N2a N2b N2c N3 Pathologic overall stage I II III IVa IVb
No. of patients (%) (n = 9) 1 (11) 6 (67) 2 (22) 0 (0) 0 (0) 5 (56) 3 (33) 1 (11) 4 (44) 0 (0) 0 (0) 4 (44) 1 (11) 0 (0) 0 (0) 2 (22) 2 (22) 5 (56) 0 (0)
AJCC indicates American Joint Committee on Cancer.
3.3. Treatment Planned tracheostomy placement was performed at the start of the procedure in 7 patients. Four patients, whose tumors obstructed the larynx, underwent tracheotomy with local anesthesia while awake. Two patients underwent standard endotracheal intubation with a laser-safe endotracheal tube. In no cases was TORS abandoned or converted to an open approach. Frozen section and final pathologic margins were clear of disease in all cases. In 4 cases (44%), the margins were cleared on frozen section with the initial resection, and 5 cases (55%) required 2 or more margin excisions. The mean number of attempts to achieve clear margins was 2.1 (median, 2; range, 1–4). Neck dissection was performed at the same time as TORS in all cases. Three patients (33%) underwent unilateral select neck dissections, and 6 (67%) underwent bilateral select neck dissections. One patient underwent bilateral neck
dissection before TORS and subsequently had difficult laryngeal exposure because of tongue base and epiglottic edema. In subsequent cases, TORS was performed before neck dissection. Three patients (33%) were treated with surgery alone; 4 patients (44%), with surgery and radiotherapy; and 2 patients (22%), with surgery and chemoradiotherapy.
Table 1 Patient characteristics
3.5. Oncologic outcomes
Characteristic
No. of patients (%)⁎ (n = 9)
Age (y), mean (range) Men Tobacco use Current user Previous user † Never used Alcohol use ≥7 drinks per wk b7 drinks per wk
61.9 (45–77) 7 (78)
⁎ Unless otherwise stated. † Quit more than 6 months before surgery.
4 (44) 3 (33) 2 (22) 4 (44) 5 (56)
3.4. Complications No acute perioperative complications occurred. In 1 patient, tracheostomy site bleeding developed 8 months after surgery, which required surgical exploration and ligation of an inferior thyroid artery. In another patient, supraglottic stenosis developed, which required prolonged tracheostomy and temporary placement of an endolaryngeal stent. The tracheostomy and endolaryngeal stents were subsequently removed, and the patient had a patent and functional airway.
The mean follow-up period was 26 months (range, 5–46 months). No patients were lost to follow-up. At last followup, 1 patient had died of disease, and 2 patients died of unrelated causes (pulmonary embolus at 5 months, myocardial infarction at 15 months). The remaining 6 patients were alive with no evidence of disease. Seven patients (77%) had complete disease control, and 2 (22%) had tumor recurrence. A patient with a stage IVa T2N2bM0 tumor treated with TORS, unilateral neck dissection, and radiotherapy had contralateral regional recurrence 9 months after TORS. The patient underwent contralateral neck dissection and died of
382
S.M. Olsen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379–384
Table 3 Disease control and survival estimates at 24 months Outcome Disease control Local Regional Distant Overall Survival Disease-specific Overall
Kaplan-Meier estimate 100% (2) 87.5% (2) 100% (2) 87.5% (2) 87.5% (2) 66.7% (2)
Values are presented as percentage (no. at risk).
uncontrolled regional disease 15 months after initial therapy. The second recurrence was in a patient with a stage III T3N0M0 tumor treated with TORS and bilateral neck dissection. A second primary tumor developed in the contralateral tonsil 21 months after TORS and was also treated with TORS. The patient had subsequent local recurrence at the supraglottic site 34 months after the original surgery, which was treated with chemotherapy and radiotherapy; the patient is currently alive with no evidence of disease 46 months after initial treatment. Kaplan-Meier 24-month disease control and survival estimates are shown in Table 3. 3.6. Functional outcomes After TORS, all patients were able to communicate verbally. Speech outcomes included normal speech in 3 patients, mild dysphonia in 2 patients, and moderate dysphonia in 4 patients. No patients had severe dysphonia. Temporary tracheostomy was required in 7 patients (78%) for airway protection during therapy. At last follow-up, 2 patients (22%), both of whom were treated with multimodality therapy, remained tracheostomy dependent. Postoperative swallow studies revealed safe swallowing ability in 5 patients (55%). Four patients (44%) had postoperative dysphagia and underwent G-tube placement. At last followup, 2 patients (22%), both of whom were treated with multimodality therapy, remained G-tube dependent. 4. Discussion Several small series have demonstrated the feasibility and short-term functional outcomes achievable when treating SG SCC with TORS. Weinstein et al [4] reported the first use of TORS for SG SCC; 3 SG SCCs were successfully resected with no operative complications and with functional laryngeal preservation. Park et al [7] also reported the successful resection of a supraglottic carcinoma without complications. A study by Boudreaux et al [6] included 9 patients with laryngeal cancer (subsite not specified), 7 of whom were successfully treated with TORS. These studies demonstrate that TORS is a feasible modality for resection of select laryngeal cancers, with
low associated morbidity and acceptable short-term functional outcomes. The study by Boudreaux et al [6], in which several tumors were unresectable via TORS, demonstrates the difficulty of proper patient selection. The patient and tumor parameters we routinely assess to determine TORS eligibility are described in the “Materials and methods” section. Some surgeons advocate performing direct laryngoscopy to assess patient and tumor characteristics and to determine eligibility for TORS. However, direct laryngoscopy adds expense, delays treatment, and adds perioperative risks. In contrast, a careful head and neck examination and radiographic studies can determine neck extension, mandible and tongue dimensions, and tumor location and extent, all of which serve to predict success with TORS. If a tissue biopsy is needed in the operating room, we perform direct laryngoscopy with biopsy and, if the results are positive, immediately proceed to TORS. To avoid the unnecessary risk and cost of placing the patient under anesthesia multiple times, direct laryngoscopy and biopsy, TORS, and neck dissection are always performed as a single procedure. Physician experience and institutional resources are important considerations when making the decision whether to perform TORS for SG SCC. Physicians must have adequate training with the robotic device, experience with endoscopic partial laryngeal surgery, and familiarity with open laryngeal surgery. Expedient and accurate frozen section pathologic analysis is also a critical component to the successful application of TORS for head and neck cancers. Communication between surgeon and pathologist is critical to ensure that margins are oriented and evaluated appropriately. The surgical techniques applied in TORS are not novel. The techniques used have been adapted from established transoral laryngeal conservation surgeries. Transoral robotic surgery builds on these proven surgical techniques with the many added benefits of robotics. In our experience, TORS can often be performed faster and often facilitates an en bloc tumor resection when compared with other transoral techniques. With robotics, the surgeon is provided magnified, 3-dimensional visualization, which allows for improved estimation of the tumor margin interface. Exposure of tumors is improved compared with that obtained with standard operating laryngoscopes. Robotics also avails the surgeon of a wide variety of instruments including electrocautery and laser options. The manual control of the robot is intuitive and provides the surgeon both tremor reduction and the ability to resect in multiple directions. In addition, TORS has the benefit of providing another “surgical assistant,” allowing up to 4 instruments to work together simultaneously. Robotics allows the physician to meticulously remove diseased tissue while preserving unaffected tissue, with the goal of retaining laryngeal function. Physicians using TORS must also understand the limitations of the device when applied to head and neck
S.M. Olsen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379–384
surgery. The primary limitation of TORS in the aerodigestive tract is the narrow oral opening that necessitates acute angles of approach between the robotic arms and the camera. As the working arms are advanced farther into the aerodigestive tract, the angles between the arms become increasingly acute (Fig. 1), thereby leading to contact interference. This problem is less prominent when performing oropharyngeal surgery but can become a major issue for laryngeal surgery in which range of motion is often severely restricted. This problem will likely be improved upon in the future with advances in robotic design and TORS retractors. To our knowledge, this is the first series to report both oncologic and long-term functional outcomes with use of TORS exclusively for SG SCC. In this series, no perioperative complications occurred, demonstrating that TORS can be performed safely. Transoral robotic surgery enabled transoral resection of tumors that may have been difficult to remove with standard transoral techniques. Importantly, the use of TORS decreased patient exposure to chemotherapy and radiotherapy. At our institution, patients
Fig. 1. Robot in position during extirpation of supraglottic tumor. Photograph demonstrating the acute angle between the robot working arms and the 30° telescope.
383
treated for supraglottic cancers with nonoperative therapy generally receive combined chemotherapy and radiotherapy for advanced-stage lesions, and radiotherapy alone for earlystage lesions. In this series, had patients been treated with nonoperative therapy, 2 patients would have received radiotherapy alone, and 7 would have received chemotherapy and radiotherapy. Based on these standard nonoperative treatment patterns, the use of TORS spared 2 patients from chemoradiotherapy, 3 patients from chemotherapy, and 1 patient from radiotherapy. As experience and skill with this technique increase, TORS may enable some patients who would otherwise require open laryngeal surgery to undergo transoral resection. Given the small patient cohort, broad conclusions about the oncologic efficacy of TORS are not possible. However, results from this series suggest that TORS achieves outcomes comparable with the oncologic outcomes reported with transoral laser microsurgery [3,8,9]. Grant et al [3] reported results of 38 patients with SG SCC treated with transoral laser microsurgery and adjuvant therapy as indicated. In that series, local control was 97%, locoregional control was 94%, and disease-specific survival was 80%—results similar to those achieved in our series. Furthermore, the 79% functional laryngeal preservation reported in that series is comparable with the high rates of voice preservation, swallowing preservation, and tracheostomy-free survival achieved in our study [3]. The functional laryngeal preservation rate achieved in our series is notable, particularly because laryngeal tumors are more likely than other head and neck subsites to require G-tube dependence after TORS [6]. In this series, 77% of patients underwent primary tracheostomy placement. As experience with the TORS technique increases, we most likely will be able to decrease the frequency of primary tracheostomy use in patients with small, nonobstructing primary tumors who do not require bilateral neck dissection. Given the limited world experience with TORS for SG SCC, we had a low threshold for use of temporary tracheostomy in this series. The major limitations of this study are the small patient cohort and relatively short follow-up. Five-year follow-up in a larger cohort of patients will be necessary to compare oncologic results of TORS with those of other accepted modalities. With these limitations noted, we believe that this study provides support for continued trials with TORS for SG SCC. This study demonstrates that TORS is a useful and safe technique for surgical extirpation of supraglottic tumors and achieves a high rate of functional larynx preservation and promising preliminary oncologic outcomes. Furthermore, the use of TORS may allow for decreased exposure to chemotherapy and radiotherapy in some patients. Although TORS offers the surgeon some unique technical advantages, we do not claim that it is a superior technique when compared with other transoral surgical options.
384
S.M. Olsen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 379–384
5. Conclusion Transoral robotic surgery is a feasible and safe modality for functional larynx preservation therapy in SG SCC. Transoral robotic surgery provides the surgeon unique advantages over traditional transoral techniques, allowing precise extirpation of supraglottic tumors. Preliminary oncologic results demonstrate acceptable disease control and survival outcomes with a high rate of functional larynx preservation, which justifies further trials with this modality. Long-term follow-up and larger series are needed to compare results of this modality with those of other accepted treatment modalities.
References [1] Silver CE, Beitler JJ, Shaha AR, et al. Current trends in initial management of laryngeal cancer: the declining use of open surgery. Eur Arch Otorhinolaryngol 2009;266:1333-52.
[2] Cabanillas R, Rodrigo JP, Llorente JL, et al. Oncologic outcomes of transoral laser surgery of supraglottic carcinoma compared with a transcervical approach. Head Neck 2008;30: 750-5. [3] Grant DG, Salassa JR, Hinni ML, et al. Transoral laser microsurgery for carcinoma of the supraglottic larynx. Otolaryngol Head Neck Surg 2007;136:900-6. [4] Weinstein GS, O'Malley Jr BW, Hockstein NG. Transoral robotic surgery: supraglottic laryngectomy in a canine model. Laryngoscope 2005;115:1315-9. [5] Solares CA, Strome M. Transoral robot-assisted CO2 laser supraglottic laryngectomy: experimental and clinical data. Laryngoscope 2007;117: 817-20. [6] Boudreaux BA, Rosenthal EL, Magnuson JS, et al. Robot-assisted surgery for upper aerodigestive tract neoplasms. Arch Otolaryngol Head Neck Surg 2009;135:397-401. [7] Park YM, Lee WJ, Lee JG, et al. Transoral robotic surgery (TORS) in laryngeal and hypopharyngeal cancer. J Laparoendosc Adv Surg Tech A 2009;19:361-8. [8] Motta G, Esposito E, Testa D, et al. CO2 laser treatment of supraglottic cancer. Head Neck 2004;26:442-6. [9] Remacle M, Lawson G, Hantzakos A, et al. Endoscopic partial supraglottic laryngectomies: techniques and results. Otolaryngol Head Neck Surg 2009;141:374-81.
