Clinical outcomes of transoral videolaryngoscopic surgery for hypopharyngeal and supraglottic cancer

Hypopharyngeal cancer (HPC) affects 0.8–1.3 per 100,000 persons per year in the US, accounting for approximately 6.5% of all head and neck squamous cell carcinomas (SCC) [1]. Unfortunately, prognosis of the patients with HPC reportedly remains the worst among all head and neck subsites, largely because the vast majority of the patients present at a locally advanced stage [2]. Since radical resection for HPC inevitably impairs laryngopharyngeal function, such as vocalization, swallowing, and breathing through the natural airway, organ-function preservation strategies have been increasingly developed, even for treatment of HPC, since the 1990s [3, 4].

Practically, there are three major options that meet the concept of organ-function preservation in the laryngopharyngeal region: radiation (RT) or chemoradiation (CRT), open partial pharyngolaryngectomy (PPL), and transoral surgery. RT or CRT has long been representative of non-surgical treatments, and concurrent CRT (CCRT), in particular, has been recognized as one of the standard therapies for advanced-staged HPC and supraglottic cancer (SGC) [5‐7]. However, intensified CCRT with a high-dose regimen results in severe long-term adverse effects including subsequent loss of function in preserved organs [8‐12]. Open PPL has also been established as a surgical organ-function preserving procedure for selected cases of early T-staged HPC and SGC [13‐15]. Although both oncological and functional outcomes of open PPL have shown to be eventually satisfactory, the surgical invasiveness associated with external incision, reconstruction procedure, and tracheostomy necessitate cautious postoperative managements and relatively long rehabilitation periods, which may make this procedure less popular.

Transoral surgery has emerged as another therapeutic option for laryngopharyngeal lesions. Because of its less invasiveness compared to CCRT regarding treatment-induced long-term toxicity and to open PPL regarding direct histological damage to the surrounding normal tissues, transoral surgery is expected to be an ideal alternative for the treatment of HPC patients. Traditionally, application of transoral surgery had been confined to early tumors in oral, oropharyngeal (except for tongue base), and glottic regions, because of the anatomically limited visualization and manipulation due to a lack of suitable optical instruments. Technological advancements in microscopic/endoscopic monitoring and surgical supporting devices have enabled development of various transoral surgical methods that can approach the hypopharyngeal and supraglottic regions, such as transoral laser microsurgery (TLM) using a microscope since the late 1990s [16‐22], and more recently, transoral robotic surgery (TORS) using a surgical robot since the late 2000s [23‐30].

Besides the above-mentioned procedures, Shiotani et al. have developed a distinct, unique, non-robotic surgical method custom-built for transoral partial pharyngolaryngectomy since the 2000s; this was subsequently renamed “transoral videolaryngoscopic surgery (TOVS)” [31‐33]. In this system, combined use of a distending laryngoscope with a rigid endoscope (videolaryngoscope) can provide a broad intraluminal field of view and a wide working space throughout the upper aero-digestive tract, which facilitates en bloc tumor resection via direct bimanual handling and application of the ready-made straight-form surgical instruments and devices. Favorable oncological outcomes and good functional results have been achieved so far by employing TOVS for T1, T2, and selected T3 cancers of the hypopharynx, supraglottis, and oropharynx [32, 33]. However, because it has not been long since this promising method was introduced, the long-term treatment outcome of TOVS remains to be validated.

The aim of this paper was to retrospectively evaluate clinical outcomes of TOVS for a cohort of patients with HPC and SGC in a tertiary referral center.

The present study revealed that the cohort of patients with HPC and SGC who underwent TOVS as an initial treatment according to our criteria had favorable oncological outcomes, even after a long-term follow-up period. Notably, those results were achieved with fairly low incidence of surgical complication and minimal postoperative dysfunction. Thus, we regard TOVS, in combination with neck dissection and adjuvant CCRT if necessary, as one of the excellent therapeutic strategies in terms of organ-function preservation for patients with HPC and SGC.

Although several transoral approaches as less invasive surgery than conventional open PPL have been developed so far, TOVS has its own advantages over other approaches in terms of practical usefulness. When compared with TLM [16‐22], TOVS has several technical advantages. First, since an endoscope lens possesses much longer depth range of focus (e.g., from 3 to 30 mm for a lens 4 mm in diameter) and wider angle of view (e.g., >110° for an above-mentioned lens) than that of a microscope, TOVS can provide a much broader surgical view in both horizontal and vertical directions compared to TLM, which helps improve recognition of anatomical orientation. Second, the field of endoscopic view is free from visual restriction due to the inner wall of the laryngeal blade that often restricts the microscopic view. Moreover, manipulation of surgical instruments is not restricted by a microscope interposed between patient and surgeon. Third, en bloc resection of primary tumor achieved by TOVS enables accurate evaluation of pathological findings, especially about margin status, tumor depth, and horizontal diameter, which cannot be assessed in tumor specimens resected blockwise by TLM. Such pathological information, together with differentiation, vascular invasion, and lymphatic invasion, are indispensable not only for judging completeness of resection but also for assessing risk of delayed neck metastasis in clinically N0 patients [46]—this underscores the importance of en bloc resection in decision making regarding additional intervention. Fourth, the NBI mode equipped in endoscopes, including the ENDOEYE FLEX LTF-S190-5 (Olympus), is available even during surgery if necessary [33].

Although indications for use of TORS using da Vinci surgical systems have recently been extended to HPC and SGC in several countries [23‐30], TORS has not been approved yet in many countries, including Japan. Instead, TOVS has been developed as a non-robotic transoral surgery in Japan. In comparison with TORS, TOVS has more practical advantages. First, since the surgeon bimanually manipulates surgical instruments directly on the real lesion, unlike TORS, the surgeon can recognize tactile sensations through those instruments; this is essential for assessing tumor invasion to the surrounding tissues and adding adequate counter-traction during the dissection procedure. Second, the cost-effectiveness of TOVS is far higher than that of TORS, because neither an extremely expensive surgical robot nor high-priced disposable equipment is required. Whereas most hospitals still cannot afford da Vinci surgical systems for TORS, the initial cost to introduce TOVS and its running costs are much lower, because most surgical instruments and devices are reusable, not originally designed for TOVS, and can be shared among other surgeries. Accordingly, TOVS can be introduced more easily than TORS in more hospitals in more countries.

Since endoscopes used in TOVS are not yet equipped with binocular vision, a three-dimensional view is not available. However, this does not really affect surgical performance because most otorhinolaryngologists/head and neck surgeons are already familiar with the two-dimensional endoscopic view. Fortunately, such a possible disadvantage compared to TLM and TORS is well compensated for by the use of high-resolution cameras. Furthermore, three-dimensional rigid endoscopes will be introduced in the near future, if necessary.

Besides TLM and TORS, a few other transoral approaches using a flexible gastrointestinal endoscope for early HPC were also reported with favorable outcomes from Japan: endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) performed by gastroenterologists [47‐50] and endoscopic laryngopharyngeal surgery performed by otorhinolaryngologists and gastroenterologists [43]. However, indication for use of these methods was confined to only patients with superficial lesions in the pharynx without lymph node metastasis (N0), excluding patients with SGC, invasive cancer, or lymph node metastasis (≥N1). Thus, distributions of disease stage in these patients, most of whom are at an incipient stage, are largely different from those of TOVS and others. In other words, indication for TOVS is very broad and ranges from superficial, small, or thin lesions (Tis) to invasive, exophytic, or bulky masses (up to T3 defined by size criteria). Notably, such wide-ranging lesions can be resected in the common setting with the same instruments; thus, high versatility is another advantage of TOVS.

A fair comparison of clinical outcomes between TOVS and other surgical approaches such as TLM or TORS seems difficult, because distributions of the disease stage, follow-up periods, and endpoint settings differ among them. However, oncological and functional outcomes of TOVS, including the previous report [33], are mostly comparable to those of TLM and TORS [16‐30]. Considering the distribution of disease stage in the patients in our study, a half of them were stage III–IV, the long-term oncological and functional outcomes appeared to be satisfactory. These results may corroborate an overall validity of this therapeutic strategy, including the criteria of indication, the principles of surgical management, and the standards of adjuvant treatments, especially about the relatively low necessity of postoperative RT.

However, it should be noted that strict observation in the follow-up period is another crucial prerequisite to achieve high LP-CSS and LRC-CSS in this cohort, because a majority of them possessed a high risk of developing second primary cancer in the pharyngolaryngeal region, even though the primary lesion was completely resected. In our cohort, although 18 patients developed one or multiple second primary cancer in the pharyngolaryngeal region, all but one was diagnosed at an early stage. Among them, 16 patients were able to preserve the larynx by repeated TOVS alone (n = 10), by following TOVS with RT for a third primary tumor with (n = 1) or without (n = 1) cetuximab, by open PPL alone (n = 2), by following open PPL with CCRT for a close margin (n = 1), or by RT alone (n = 1); in contrast, the other two patients ultimately required total laryngectomy for unavoidable reasons. Therefore, close attention must be maintained throughout follow-up so that second or later primary lesion can be treated appropriately as early as possible.

In both CSS and OS, advanced N stage (N2-3) was found to be the only independent unfavorable prognostic factor in this cohort, probably in part because of relatively low statistical power due to small sample size. Intriguingly, in accordance with the LRC-CSS rate as high as 88.0% at 3 and 5 years, uncontrolled regional failure and related death occurred in only one of 10 patients who developed treatment failure. The remaining nine patients developed distant metastasis without locoregional failure, seven of whom died as a consequence, suggesting that N2-3 is a strong predictor of death due to delayed distant metastasis. In agreement with these results, N2-3 was also found to be an independent unfavorable predictor of distant metastasis-free survival (data not shown). Thus, in common with many other cancers at an advanced stage, the most critical unsolved issue appears to be management of distant metastasis, irrespective of differences in therapeutic modality for locoregional lesions.

Regarding postoperative management-related issues, the incidence of temporary tracheostomy is relatively low in cohorts who underwent TOVS, including both the present (2.8%) and previous (6.7%) reports [33], while it varies largely in each of the other surgical approaches such as TLM, TORS, and ESD/EMR. Several recent studies reported a relatively high incidence of tracheostomy in TLM (12.2–16.0%) [21, 22], TORS (23.8–100%) [26, 29], and ESD/EMR (16.3%) [49]; however, these numbers seem to reflect the prophylactic use of tracheostomy to a certain extent. On the other hand, in some studies in which no patients underwent tracheostomy, instead, a high incidence of prolonged intubation of more than 24 h was reported in TLM (27.1%) [16], TORS (60.0%) [30], and ESD (30.8%) [47]. Although the necessity of tracheostomy principally depends on the extent of postoperative laryngeal edema based on depth and width of the defect after tumor resection and its location, because such estimation involved personal experience and expertise in airway management, making a decision of tracheostomy is rather subjective. In our experience, careful observation of the surgical site in view of the whole laryngopharynx under endoscope by skilled otorhinolaryngologists just after resection is sufficient to make an appropriate decision.

Fortunately, no patients experienced major postoperative hemorrhage that required return to the operation room for emergency treatment. Since we have been well aware of the potential risk of hemorrhage that can be fatal, maximum attention has always been paid to completeness of hemostasis after resection. In our practice, use of a super long bipolar forceps and/or BiClamp LAP forceps can efficiently control any active bleeding during surgery without difficulty. In case a patient has taken an anticoagulant agent, proper management of the drug during the perioperative period is also imperative to prevent increased hemorrhage.

Although aspiration pneumonia was found only in three patients, slight mucus influx into the glottis was observed temporarily in some other patients, especially at an early period after TOVS, suggesting that silent aspiration might occur more frequently. It can be assumed that such transient aspiration that could develop into pneumonia was preventively resolved in many ways, including minute endoscopic evaluation of swallowing function before and after resuming oral intake, timely removal of feeding tube without delay, thoughtful adjustment of form of meal by a dietitian, and appropriate introduction of swallowing rehabilitation if necessary. In addition, effects of preventive efforts to avoid progressive stricture, such as injection of steroid just after resection and repeated balloon dilation if necessary, also appeared to be reflected by a low incidence of persistent dysphasia. Furthermore, in some patients with advanced N stage who underwent neck dissection, reduced intensity of adjuvant CCRT, which was based on the detailed histopathological finding of lymph node metastasis, was also assumed to partly contribute to better function-preservation.

TOVS for patients with HPC and SGC as an initial treatment provided favorable long-term oncological outcomes with low frequency of surgical complication and minimal functional impairment, corroborating its validity as a therapeutic strategy for this cohort. While high LP-CSS and LRC-CSS reflected its excellent locoregional control, advanced N stage determined as an independent prognostic factor in both CSS and OS indirectly reflected higher risk of delayed development of distant metastasis as an unsolved issue. Considering its sound clinical outcomes and various practical advantages, TOVS can be a dependable, less invasive, and cost-effective surgical option of an organ-function preservation strategy for HPC and SGC.