eTable 1. List of Total Laryngectomies
eTable 2. Summary of Published SCPL Series
Sperry SM, Rassekh CH, Laccourreye O, Weinstein GS. Supracricoid Partial Laryngectomy for Primary and Recurrent Laryngeal Cancer. JAMA Otolaryngol Head Neck Surg. 2013;139(11):1226-1235. doi:10.1001/jamaoto.2013.4990
Copyright 2013 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Supracricoid partial laryngectomy (SCPL) is an essential technique in the armamentarium of modern laryngeal organ preservation surgery.
Objective, Design, Setting
Retrospective case series to review the oncologic outcomes following SCPL in a large US-based cohort treated by a single surgeon in a tertiary-care university hospital.
A total of 96 consecutive patients with primary or recurrent squamous cell carcinoma of the larynx undergoing SCPL from 1992 to 2010.
Supracricoid partial laryngectomy surgery.
Main Outcomes and Measures
Five-year local control and laryngeal preservation, using the Kaplan-Meier method.
There were 54 primary laryngeal carcinomas and 42 previously treated with radiation to the larynx; 23% were supraglottic or transglottic tumors (n = 22). The overall 5-year local control rate for the series was 94%. For T2 and T3 primary tumors, the 5-year local control was 100% and 96%, respectively. In patients previously treated with radiation, the 5-year local control was 89%, with an 89% laryngeal preservation rate. Among stage III or IV primary laryngeal tumors for which concurrent chemoradiation was a treatment alternative, the 5-year local and locoregional control was 96% and 83%, respectively, and the 5-year larynx preservation was 91%. Ultimate local control was achieved for all patients in the series. A significant postoperative complication occurred in 19% (n = 18) and 1 anesthesia-related perioperative death occurred. No total laryngectomies were performed for laryngeal dysfunction.
Conclusions and Relevance
This series demonstrates excellent local control for both primary and recurrent laryngeal cancers, with functional larynx preservation. In appropriately staged and selected patients with T2 or T3 primary laryngeal cancer or laryngeal cancer following prior radiation treatment, SCPL should be considered as a treatment alternative to non-surgical treatment or total laryngectomy.
In 1959, Majer and Rieder1 described a partial laryngectomy, which included an impaction of the hyoid to the cricoid. The group of partial laryngectomy procedures subsequently developed in France and other European countries involving this method of laryngeal reconstruction was codified under the name supracricoid partial laryngectomy (SCPL) and includes resection of the true and false vocal folds, both paraglottic spaces, and the thyroid cartilage. It optionally includes up to 1 arytenoid cartilage, a portion or all of the epiglottis, and the preepiglottic space. For supraglottic and transglottic cancer, SCPL involves resection of the entire epiglottis and preepiglottic space, and the reconstruction is a cricohyoidopexy (CHP). For glottic cancer, SCPL may preserve the epiglottis, and the reconstruction technique is a cricohyoidoepiglottopexy (CHEP).
The SCPL provides a surgical organ-preserving alternative to total laryngectomy (TL) for selected endolaryngeal malignant neoplasms. The advantage of the SCPL over TL is the preservation of lung-powered speech and swallowing function, without a permanent stoma, while offering a very high local control rate for selected laryngeal cancers. Local control rates for T2 and T3 laryngeal carcinomas of 90% and higher have been reported, which are better than those reported for radiation therapy plus chemotherapy and other forms of partial laryngectomy.2- 7 It has also been reported that SCPL was an effective treatment alternative to TL for salvage surgery following primary irradiation failure in a relatively small series of patients.8- 10
As with any technique, the key to successful outcome following SCPL is careful patient selection. The tumor must be contained within the limits of the en bloc resection area. Clear contraindications include a fixed arytenoid, subglottic extension to the level of the cricoid, hyoid invasion, and extralaryngeal spread. Maintenance of laryngeal function requires the preservation of at least 1 cricoarytenoid unit (arytenoid cartilage, cricoid cartilage, posterior and lateral cricoarytenoid muscles, and superior and recurrent laryngeal nerves). Unlike other partial laryngectomies (eg, extended vertical partial laryngectomy) that have a plethora of disparate reconstructive options, the SCPL–CHEP and SCPL–CHP involve single, fixed methods of reconstruction, which ensures reproducible functional outcomes and enhances teachability and hence generalizability of these techniques among head and neck surgeons.
In this study, we report the oncologic and surgical outcomes in a large US-based series of patients treated with SCPL by a single surgeon (G.S.W.).
The study was approved by the institutional review board of the University of Pennsylvania prior to data collection. The need for informed consent was waived. The medical records of all patients who had a diagnosis of squamous cell carcinoma (SCC) from a laryngeal primary site and underwent an attempted SCPL by the senior author (G.S.W.) between 1992 and 2010 were retrospectively reviewed. The procedure was performed as previously described, and the patients underwent reconstruction with either CHP or CHEP.11,12 When indicated, a unilateral or bilateral neck dissection was performed. All patients had a temporary tracheostomy; most patients had a percutaneous gastrostomy tube placed preoperatively; patients who declined this had a nasogastric tube placed postoperatively. Patients who underwent SCPL and then subsequently underwent intraoperative conversion to a TL because of extent of disease were included in the series data collection, but excluded from the analysis of oncologic outcomes.
The following data were collected on patients from the office notes, hospital charts, operative notes, and pathology reports: demographics, primary site, prior treatments, stage, extent of operation, pathologic characteristics, adjuvant treatments, recurrence, overall survival (OS), and complications. Stage of the disease was defined using the seventh edition of the American Joint Committee on Cancer (AJCC) staging manual based on the clinical examination findings and final pathology report. Recurrence was defined as local, regional, and distant. Salvage treatments (if attempted) and outcomes were documented. Public death records were searched to determine dates of death. Local control (LC) was defined by the time interval between the date of operation and documentation of recurrence at the site of the primary tumor. Local recurrences were distinguished from second primary tumors arising at adjacent sites only if (1) the sites were clearly completely separated by normal intervening tissue or (2) sufficient time (ie, >5 years) had elapsed between the tumor occurrences. Patients who had not developed recurrence were censored at the time of last follow-up visit, regardless of regional and/or distant recurrence. For disease-specific survival (DSS) analysis, patients who were free of disease were censored at their last follow-up visit. For larynx preservation (LP), patients without local recurrence were censored at their last follow-up if they had not had a laryngectomy performed for control of the primary disease or aspiration. Local control, DSS, LP, and OS were calculated by the Kaplan-Meier method. All analyses were done using SPSS software, version 20 (SPSS Inc).
A list of the demographic, clinical, and pathologic features of the series is given in Table 1. For comparison purposes, the series was further subdivided by prior treatment with radiotherapy to the larynx; 44% of the cohort had previously been irradiated (n = 42). This group of patients differed significantly from the primary cases in terms of laryngeal subsite involved, T stage, N stage, overall stage, depth of invasion, paraglottic space involvement, preepiglottic space involvement, the extent of neck dissection, and the occurrence of distant metastases. However, perineural invasion (PNI), lymphovascular invasion (LVI), the ability to achieve negative margins, and recurrences locally and regionally were not significantly different between those previously treated with radiation and those not treated with radiation.
Neck dissection treatment and nodal staging are summarized in Table 2 and Table 3, respectively. The patterns of lymphatic spread were consistent with reported literature. Regional metastases did not occur among recurrent laryngeal tumors that were previously irradiated. Regional metastases occurred only in T3 and T4 primary tumors, and mainly among the supraglottic primaries.
Eighty-five of the 96 patients had no further evidence of disease following SCPL. A summary and a clinical case description of the 11 failures is listed in Table 4. Local recurrence was observed in 5 patients, only 1 of whom was in the primary tumor group. This patient had successful salvage total laryngectomy but later died of distant disease. All 4 patients with local recurrence following salvage SCPL had their disease successfully controlled with total laryngectomy. The remaining failures occurred among the primary tumor group, and were regional (n = 3) or distant (n = 3).
The oncologic outcomes are summarized in Table 5. The overall LC (prior to salvage surgery) for the series at 5 years was 94%. For primary laryngeal tumors not previously treated with radiation, the 5-year LC was 98%, and for T2 and T3 tumors, this was 100% and 96%, respectively. For T2 primary laryngeal tumors, 100% were able to preserve their larynx at 5 years, and for T3 primary tumors the 5-year LP was 91%. In cases of SCPL performed for tumors in previously irradiated fields, the overall 5-year LC (with organ preservation) was 89%; 55% of these tumors were rT1 (n = 23), and the 5-year LC was 91%. Overall, 89% of these patients with recurrent laryngeal tumors were able to preserve their larynx. The ultimate LC in this series was 100% because all local failures were successfully salvaged with total laryngectomy to control the primary site disease. A summary of the total laryngectomies performed is provided in eTable 1 in the Supplement.
Four patients had positive margins following SCPL—3 previously mentioned were cases of irradiation: 2 had grossly positive margins intraoperatively precluding SCPL and were converted to TL with eventual negative margins; 1 had microscopically positive margins on final pathology evaluation, which was treated with a completion laryngectomy and adjuvant chemotherapy and irradiation. The fourth patient had a T3 supraglottic primary tumor, with microscopically positive margins on final pathologic analysis, which was treated with adjuvant irradiation. There were no recurrences in this group of patients with positive margins. Three patients had close margins seen on final pathologic findings (<2 mm from inked border); 1 of these patients received additional adjuvant treatment, and no recurrences developed in this group. Seven patients had high-grade dysplasia or carcinoma in situ at the surgical margin; none of these patients received adjuvant treatment or developed recurrences.
Twelve patients (12%) were treated with adjuvant radiation postoperatively. One of these was a case of irradiation failure with microscopically positive margins, and the patient underwent a TL followed by adjuvant reirradiation treatment. The other 11 patients had primary laryngeal tumors (20%) and were treated with postoperative radiation therapy for multiple positive neck nodes (n = 5), microscopically positive margins (n = 1), close margin (n = 1), LVI (n = 1), PNI (n = 1), thyroid cartilage invasion (n = 1), and advanced stage (n = 1). One patient had extensive LVI and a preexisting tracheostomy without positive nodes and was treated with adjuvant irradiation; despite the adjuvant treatment, this patient developed a stomal recurrence and distant metastases. None of the other patients treated with adjuvant radiation therapy had a locoregional recurrence. Findings in patients with indications for postoperative adjuvant radiation therapy but who did not receive radiation therapy included multiple positive neck nodes and extracapsular spread (n = 1) and LVI (n = 5). One of these patients with a T3N0 glottic primary with LVI who did not receive adjuvant radiation therapy developed a local, regional, and distant recurrence. Patients did not undergo indicated radiation because of patient choice or delay secondary to medical issues or surgical complications.
Significant surgical complications of the procedure occurred in 19% of cases (n = 18), which led to hospitalization or additional treatment. The complications included ruptured pexy (n = 2), wound infection/fistula (n = 4), aspiration/tracheitis (n = 4), chyle leak (n = 4), Clostridium difficile colitis (n = 2), and alcohol withdrawal (n = 1). There was 1 anesthesia-related perioperative death early in the series due to hepatic failure secondary to the toxic effects of 2 consecutive doses of halothane, one at the time of a panendoscopy and the second at the time of the SCPL. The median length of hospitalization was 6 days.
All patients were discharged with the tracheostomy tube in place. The median duration of the tracheostomy tube remaining in place was 43 days. All but 6 patients were decannulated following SCPL: 2 were lost to follow-up prior to decannulation; 3 had recurrent or second primary disease, which precluded decannulation; and 1 developed a stricture. No total laryngectomies were performed for laryngeal dysfunction.
All patients were discharged home with feeding tubes. Swallowing data were available for 93 of the 96 patients; 3 patients had limited follow-up or documentation, which precluded retrospectively determining swallowing outcome. We consider 1 patient an outlier, who initially had the feeding tube removed following surgery and then replaced 3 months later, and kept it for 5 more years until it was removed again. This patient had complicated medical issues, which led to intermittent periods of dysphagia and feeding tube use during ongoing treatment. A total of 87 patients (94%) had their feeding tubes removed during follow-up (range of feeding tube placement duration, 22-792 days; median duration, 74 days). One of these patients could functionally swallow but chose to retain and use the feeding tube because the patient believed that enteral nutrition improved their chronic pancreatitis condition. Four of these patients had persistence of the cancer and additional treatments, including radiation therapy and chemotherapy. Only 1 patient retained the feeding tube because of persistent difficulty swallowing and aspiration; this patient had a recurrent T2N0 glottic tumor and had been previously irradiated to 72 Gy. Excluding patients who had tumor recurrences requiring further oncologic treatment following SCPL, the rate of dependence on a percutaneous endoscopic gastrostomy (PEG) tube was 1% (1 of 75); and among primary laryngeal tumors, there were no patients dependent on a PEG tube following SCPL.
This report is to our knowledge the largest series description of SCPLs performed in the United States to date. This procedure is widely performed across the world, and series have been published from 4 different continents.5 However, relatively few series have been reported from the United States.13 The results for SCPL from the current series are consistent with the outcomes reported in series across the world. The reported 5-year LC ranges from 83% to 97% in various series.5 The weighted-pooled LC rate from these series is 93%, representing 1400 patients total, where 60% were T2 and 23% were T3 tumors (eTable 2 in the Supplement). In this current series we report an overall 5-year LC of 94%, where T2 and T3 tumors represent 70% overall. If we exclude the previously irradiated larynges and count only the primary tumors in this series, the 5-year LC was 98% among T2 or T3 tumors. It is clear that this series is consistent with the outcomes reported in the rest of the world among at least 1400 patients treated with SCPL, with a 5-year LC average close to 93% for T2 and T3 tumors.
In 2013, 2 potential alternative surgical larynx-preserving treatments for laryngeal cancers may include transoral laser microsurgery (TLM) or transoral robotic surgery (TORS). TLM for primary T2 glottic tumors has a local recurrence rate of 7% to 26% at 5 years in various studies,14 which is inferior to the 0% local recurrence we report for T2 glottic primary tumors. In addition, there is no compelling evidence in the literature that the extensive TLM resection required for T2 glottic carcinoma leads to a superior voice outcome compared with SCPL. TORS is used for supraglottic tumors that do not extend to the level of the glottis and for which the initial reports of oncologic and functional outcomes are promising15; whereas SCPL is used for supraglottic tumors that extend to the glottic level. Therefore, the techniques of TORS and SCPL are complementary, and the indications do not overlap.
The alternative nonsurgical larynx-preserving treatments for the patients with primary laryngeal tumors not previously irradiated was either radiation therapy alone or with concurrent chemotherapy. Considering specifically the T2 primary laryngeal tumors (n = 23), the 5-year LC and LP in this series was 100%. Comparing this to the reported 5-year LC for T2 tumors treated with radiation alone, the rate ranges in the literature from 50% to 85%, with LP of 80% and DSS of 90%.16- 19 So for the 23 patients with T2 primary tumors in this series, we would predict that a minimum of 3 would have had a local recurrence and likely undergone a TL had they been treated with primary irradiation, which they avoided when treated with SCPL. All of the patients with T2 primary tumors treated in this series retained a functional larynx with normal swallowing function and no feeding tube dependency. As recognized in the 2006 American Society of Clinical Oncology(ASCO) practice guidelines for larynx cancer, patients with T2 laryngeal cancers should be counseled that laryngeal preservation following SCPL is superior to radiation therapy with otherwise good functional outcomes—with the caveat that the patient should be willing to accept worse hoarseness following SCPL.20 As pointed out in the ASCO practice guidelines, this recommendation is based on the higher local failure with primary radiotherapy for T2 tumors compared with SCPL. When disease recurs following radiation therapy and requires a total laryngectomy, the 5-year survival is less than 50%.21 This low survival rate following laryngectomy for irradiation failure is of great significance, since only 1 in 3 of the patients for whom radiation therapy fails will still be amenable to an organ preservation surgery.22,23 Indeed, all of the patients with T2 laryngeal cancers in this series were presented with both options and chose to undergo SCPL rather than irradiation because of the better predicted outcomes in terms of overall LP and survival despite the fact that all of the patients were counseled concerning the outcome of significant hoarseness from the procedure.
T2 primary glottic tumors are the ideal and most common indication for SCPL. However, even in one of the busiest centers in the United States, approximately 5 of these tumors is treated with SCPL per year. The annual incidence of T2 glottic primaries in the United States was estimated from SEER data in 2009 to be 1000 (personal communication, Amy Chen, MD, Emory University, 2013). The incidence of larynx cancer is on the decline, so this number of T2 glottic cancers in the United States is likely to be even lower in the future.24 If we predict that 50% to 90% of patients would opt for SCPL if given the option, the annual number of cases nationally would be between 500 and 900. Furthermore, if we estimate that there are currently from 50 to 150 surgeons in the country who perform SCPL, there would be from 3 to 18 T2 glottic primary lesions treated per year per surgeon. This volume is well within the capacity of currently trained head and neck surgeons. Although learning curve results for SCPL have not been reported, to our knowledge, the senior author (G.S.W.) has observed during a 22-year experience at the University of Pennsylvania that even limited exposure to SCPL by residents and fellows has been sufficient training to allow those with an interest to perform these procedures. We hypothesize, however, that the limited exposure to the SCPLs, both during training and in practice, by most general otolaryngology–head and neck surgeons, who most often function as the gatekeeper for patient management decisions, may result in patients with T2 glottic carcinomas not being offered a second opinion with surgeons who have the ability to assess whether SCPL is indicated for a particular patient. Data from the National Cancer Database showed that early-stage glottic cancer is less likely to be treated with surgery and more likely to have a worse survival when treated outside of academic teaching hospitals.25 Outcomes data discussed herein suggest that many patients with T2 glottic tumors might prefer SCPL to radiation therapy because of higher local control and likelihood of LP in addition to other factors. Therefore, it is important that these patients are evaluated by surgeons with expertise in organ preservation surgery generally, and SCPL specifically, so that they may be counseled appropriately about the options and participate in the treatment decision making.
Among the 29 patients with advanced-stage disease who would have been offered concurrent chemoradiation as the alternative nonsurgical LP treatment, the 5-year LC and locoregional control rates were 96% and 83%, respectively, and the 5-year LP rate was 91% in this series. All of these patients had T3 tumors, except 1 with T4 disease, based on pathologic staging. The decision to operate on T3 tumors must be made with an understanding of the contraindications to the SCPL, which have been clearly delineated previously.26 A T3 tumor with vocal cord fixation (which implies paraglottic space invasion) is a candidate for the SCPL as long as the arytenoid is mobile, which can be assessed at office endoscopy. A fixed arytenoid indicates tumor invasion of the cricoarytenoid joint, and it is not a candidate for the SCPL. The pT4 patient was clinically staged T3 and had extension of the tumor through the thyroid cartilage at final pathologic analysis, which would generally be a relative contraindication to the procedure. Thyroid cartilage invasion alone does not predict worse prognosis, but invasion through the thyroid cartilage outer perichondrium does worsen survival.27,28 Although some have reported cases of T4 tumors successfully treated with SCPL by not dissecting the strap muscles off of the larynx, this is not recommended for the novice.29 Half of the advanced-stage tumors in this series were supraglottic; one-third had neck disease; and one-third were treated with adjuvant irradiation. Two patients had significant surgical complications, including a fistula and a chyle leak, which they recovered from. Excluding patients who retained a tracheostomy and PEG tube during ongoing treatment with irradiation and chemotherapy for recurrences, all patients were decannulated except 1, who developed a stricture; all patients had PEG tubes removed. The most likely cause of failure in this group was from distant metastases, which occurred in 5 patients (17%). There was only 1 local failure, successfully salvaged with a total laryngectomy, and 4 regional failures.
The oncologic outcomes for the advanced-stage patients in this series appear to be better than that reported previously for nonsurgical organ-preserving treatment with concurrent chemoradiation, where the 5-year LC, locoregional control, and LP rates were 71%, 68%, and 84%, respectively.4,30 In the Radiation Therapy Oncology Group (RTOG) 91-11 trial,31 treatment with combined chemoradiation resulted in a 43% rate of severe late complications, including an 18% rate of pharyngeal dysfunction and a 25% rate of laryngeal dysfunction. In addition, there was a 7% rate of death due to the treatment protocol and a high rate of complications following salvage TL.30,32 Thus, we interpret the available data to indicate that between these 2 organ-preserving treatment approaches for advanced-stage laryngeal cancer, SCPL gives the more favorable rate of local oncologic control and functional outcome for those selected patients who do not have contraindications. Without a randomized clinical trial comparing surgical and nonsurgical organ-preserving treatments, observational studies, with their associated limitations, remain the best level of evidence for surgical outcomes. Appropriate treatment must be patient and disease specific in each case, and treatment decisions must be made in conjunction with multidisciplinary care. It is clear from our data that for a subset patients with advanced-stage laryngeal cancer, organ-preserving surgical options exist, and the alternative to nonsurgical organ-preservation treatment is not only a TL. The indications for SCPL are well delineated in the literature, while the indications for concurrent chemoradiation are still being defined. Nonsurgical organ preservation is appropriate for a limited subset of patients with advanced-stage laryngeal cancer, with questions remaining about the appropriate treatment of T3-fixed cord lesions and T4 tumors; TL is appropriate for a large proportion of advanced-stage laryngeal tumors.33 It is the standard of care that an informed discussion of both surgical and nonsurgical organ-preserving treatments be conducted with the patient, when the patient is a candidate for multiple treatment options, and that the patient be allowed to participate in the decision process leading to the most appropriate treatment.34
This SCPL series is unique in including a large group of patients previously treated with radiation to the larynx. Most of these patients were not treated with chemotherapy in addition to radiotherapy. Local recurrences of laryngeal tumors after primary radiation treatment are not uncommon, ranging from 5% to 15% of T1a, 7% to 18% of T1b, and 15% to 31% of T2 tumors.16- 19 Historically, these recurrent tumors after irradiation failure have been treated by several different approaches, including TLM, external partial laryngectomy approaches, or TL.35 The LC rates after endolaryngeal laser approaches have been reported to range from 50% to 85% in various series (see eReferences in the Supplement).36 Ramakrishnan et al37 summarized 11 articles including 286 patients who underwent TLM for radiorecurrent tumors. The pooled LC was 57% after the first TLM procedure, 64% after a second TLM procedure, and 88% after a salvage TL. Of open partial laryngectomy approaches for irradiation salvage, the most abundantly reported approach has been the vertical partial laryngectomy: in series totaling 462 patients combined, local control ranging from 55% to 100% with a weighted average of 77% have been reported (see eReferences in the Supplement).36 SCPL for irradiation salvage has been studied less frequently; Paleri et al38 summarized 8 series describing a total of 149 patients worldwide who underwent SCPL for radiorecurrent tumors (see eReferences in the Supplement). The local control in these series ranged from 70% to 100%, with a pooled estimate of 94%. The 42 patients in the present report represent the largest US cohort treated with SCPL for irradiation salvage. Three of these cases were converted to TL for positive margins after attempted SCPL. In the other 39 cases, the 5-year local control was 89%. Four patients (10%) developed local recurrences, which were successfully salvaged by TL, for an ultimate disease control of 100%. The 5-year OS for the patients with recurrences following irradiation treated with SCPL was 95%. Therefore, this study demonstrates that for selected patients with local recurrences at the primary site following radiation treatment, SCPL offers an outstanding chance for local control while preserving lung-powered speech and swallowing. Failures following SCPL can still be successfully salvaged with TL without compromising oncologic disease control.
Significant complications in the previously irradiated patients undergoing open neck surgery were not increased over those who had not previously been irradiated (about 20% overall; n = 8 of 39). Irradiated patients had a slightly increased incidence of pharyngocutaneous fistula (5% vs 0%) and chyle leak (8% vs 2%). Patients in both groups were decannulated equally successfully. The average duration of tracheostomy tube use was actually shorter for the previously irradiated group (52 vs 72 days, not statistically significant) likely because more patients with primary tumors underwent adjuvant irradiation, in which case they kept the tracheostomy tube in until the completion of radiation treatment. One patient who had previously been irradiated retained the PEG tube for swallowing dysfunction; overall, patients in both groups were equally successful in swallowing. The length of hospitalization was also not different for the previously irradiated patients (average, 6.5 days in both groups). Therefore, we conclude that the surgical outcomes after SCPL are not significantly affected by prior radiation treatment to the larynx, and the procedure is safe in these patients.
This study is subject to the limitations of retrospective cohort studies, including inherent biases in the retrospective data collection methods. Single-arm observational cohorts are also potentially subject to selection biases. This bias was minimized by offering all patients who were candidates for the SCPL procedure, from an oncologic and functional standpoint, the option of SCPL vs nonsurgical alternatives. In addition, as the procedure has been used for nearly 50 years worldwide, the selection criteria are well established and did not vary over the course of the study period. Therefore, this observational cohort represents a relatively homogeneous group treated by a single surgeon with minimal selection bias.
Even in the busiest center, the SCPL procedures are performed infrequently, making other methods of study, including randomized or prospective trials, impractical. Although some clinical researchers would suggest that all therapeutic interventions must be validated by randomization, it is the contention of the present authors that when retrospective, single-arm observational trials remain the best possible level of evidence regarding the safety and efficacy of a particular therapeutic intervention, it is the responsibility of clinicians to make judgments based on the best available evidence. In fact, the use of randomization to answer head and neck oncologic questions, studies in which one arm is a surgical approach and the other is nonsurgical, is fraught with inherent difficulties. Indeed it is possible that attempts to generalize findings from randomized trials that are designed to assess the optimal nonsurgical regimen for the management of laryngeal cancer may have led to decreased survival trends for advanced laryngeal cancer noted in studies using national cancer registries both in Canada and the United States.39,40 The present report is valuable both in the size of the cohort reported and the large number of patients who had previously been treated with radiation for their laryngeal primary tumors. This study, together with the large number of other retrospective SCPL studies, should be used by both head and neck surgeons and general otolaryngologists to advise their patients about the potential value of SCPL as an organ-preserving treatment option available for primary and recurrent laryngeal cancers.
In conclusion, this large series of patients treated with SCPL demonstrates excellent LC and functional outcomes for both primary and recurrent laryngeal tumors. In appropriately staged and selected patients with T2 or T3 primary laryngeal tumors, or early to moderately advanced laryngeal tumors following prior radiation treatment, SCPL should be discussed in the treatment algorithm for organ-preserving laryngeal treatment.
Corresponding Author: Gregory S. Weinstein, MD, Department of Otorhinolaryngology–Head and Neck Surgery, Center for Head and Neck Cancer, Hospital of the University of Pennsylvania, 3400 Spruce St, 5 Ravdin Bldg, Philadelphia, PA 19104 (Gregory.Weinstein@uphs.upenn.edu).
Submitted for Publication: February 8, 2013; final revision received July 17, 2013; accepted August 8, 2013.
Published Online: October 3, 2013. doi:10.1001/jamaoto.2013.4990.
Author Contributions: Drs Sperry, Rassekh, and Weinstein contributed equally. Dr Sperry had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Sperry, Rassekh, Laccourreye, Weinstein.
Acquisition of data: Sperry, Weinstein.
Analysis and interpretation of data: Sperry, Rassekh, Weinstein.
Drafting of the manuscript: Sperry.
Critical revision of the manuscript for important intellectual content: Sperry, Rassekh, Laccourreye, Weinstein.
Statistical analysis: Sperry.
Administrative, technical, or material support: Weinstein.
Study supervision: Rassekh, Laccourreye, Weinstein.
Conflict of Interest Disclosures: Dr Weinstein receives royalties from Olympus Corporation. No other conflicts of interest were reported.
Previous Presentation: This study was presented at the American Head and Neck Society 2013 Annual Meeting; April 11, 2013; Orlando, Florida.
Additional Contributions: Amy Chen, MD, provided written permission to cite her personal communication in this report.