Salvage Circular Laryngopharyngectomy and Radial Forearm Free Flap for Recurrent Hypopharyngeal Cancer

Transcription

1 The Laryngoscope VC 2013 The American Laryngological, Rhinological and Otological Society, Inc. Salvage Circular Laryngopharyngectomy and Radial Forearm Free Flap for Recurrent Hypopharyngeal Cancer Nicolas Fakhry, MD; Emmanuel Chamorey, PharmD, PhD; Justin Michel, MD; Charles Collet, MD; Laure Santini, MD; Gilles Poissonnet, MD; Jose Santini, MD; Patrick Dessi, MD; Antoine Giovanni, MD, PhD; Olivier Dassonville, MD; Alexandre Bozec, MD, PhD Objectives/Hypothesis: To evaluate surgical and long-term oncologic outcomes in a series of patients who underwent circular total laryngopharyngectomy with tubed radial forearm free flap as salvage surgery. Study Design: Retrospective study in two tertiary referral centers. Methods: Twenty-one patients who underwent surgery between 2001 and 2010 were included in the study. All patients underwent surgery for recurrence of advanced hypopharyngeal squamous cell carcinoma previously treated by chemoradiotherapy (CRT) or surgery followed by postoperative radiotherapy or CRT. Overall survival (OS) and disease-free survival (DFS) were determined by Kaplan-Meier analysis. The search for parameters that could influence long-term oncologic outcomes was carried out by univariate and multivariate analysis using log-rank test and Cox regression models. Results: Median follow-up was 30 months. The 1, 2, and 5-year OS rates were 46%, 40%, and 16%, and the DFS rates were 42%, 30%, and 15%, respectively. Free flap failure (P [log rank] ¼.014 and P [Cox] ¼.016) and positive margins (P [log rank] ¼.001 and P [Cox] ¼.001) were found to have a significant negative impact on both OS and DFS in both univariate and multivariate analysis. Conclusions: Salvage surgery remains the only curative option as treatment of recurrent hypopharyngeal squamous cell carcinoma and should be offered to patients when possible. However, selection of patients eligible for this type of surgery and the choice of reconstructive procedure are crucial since long-term survival is clearly correlated with the presence of clear margins and the absence of free flap failure. Key Words: Salvage surgery, free flap, head and neck cancer, total pharyngolaryngectomy, hypopharynx, reconstruction. Level of Evidence: 4. Laryngoscope, 123: , 2013 INTRODUCTION Organ preservation protocols with a combination of chemoradiation (CRT) are currently accepted as the firstline therapeutic modality for patients with locally advanced hypopharyngeal squamous cell carcinoma From the Service d ORL et de Chirurgie Cervico-Faciale (N.F., J.M., C.C., L.S., P.D., A.G.), Centre Hospitalier Universitaire La Timone, Marseille, France; Department of otolaryngology and head and neck surgery, University Teaching Hospital of La Timone, Marseille, France, Centre Antoine Lacassagne (CAL)/Institut Universitaire de la Face et du Cou Comprehensive Cancer Center of Antoine Lacassagne (CAL)/Head and Neck Universitary Institute (IUFC) - Nice Sophia Antipolis University, Nice, France (IUFC) Universite Nice Sophia Antipolis (E.C., G.P., J.S., O.D., A.B.), Nice, France; and Groupe Regional de Reflexion en Oncologie Cephalique (GRROC) GRROC Regional group of Head and Neck Cancer Research (N.F., G.P., A.G., O.D., A.B.), Nice, France. Editor s Note: This Manuscript was accepted for publication September 17, Presented at the 44th Annual Meeting of the SFCCF: Societe Française de Carcinologie Cervico-Faciale (French Society of Head and Neck Cancer), Besançon, France, November 25 26, The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Nicolas Fakhry, MD, Assistance Publique-- Hôpitaux de Marseille (AP-HM)/Aix Marseille Univ, Centre Hospitalier Universitaire (CHU) la Timone Service ORL et Chirurgie Cervico-Faciale, 264, rue Saint Pierre, Marseille cedex 05, France. nicolas.fakhry@ap-hm.fr DOI: /lary (SCC). 1 The management of recurrences after such treatments remains challenging, particularly because of postradiation inflammation and fibrosis. 2 6 When surgical extirpation is possible, treatment generally requires a total laryngopharyngectomy. When tumor extends circumferentially to the whole hypopharyngeal mucosa or involves the pharyngoesophageal junction, a circular total laryngopharyngectomy (CTLP) with reconstruction becomes necessary. Several reconstructive procedures from regional pedicled flaps to free flaps have already been described. 5 Moreover, many reports have been published concerning the different reconstruction techniques and their postoperative course. To date, however, few recent studies have focused on oncologic results in these patients. 4,7 9 The aim of this study was to evaluate surgical and long-term oncologic outcomes in a series of patients who underwent CTLP with tubed radial forearm free flap (RFFF) as salvage surgery for recurrence of advanced hypopharyngeal SCC previously treated by CRT or surgery followed by postoperative radiotherapy (RT) or CRT. MATERIALS AND METHODS Population We conducted a retrospective study including all patients (N ¼ 21) who underwent salvage CTLP and reconstruction with tubed

2 Fig. 1. Circular total laryngopharyngectomy with radial forearm free flap reconstruction. (A) Flap raising. (B) Vascular microanastomosis. (C) Flap setting. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] RFFF in two tertiary referral centers (a university teaching hospital and a comprehensive cancer center) between 2001 and Another four patients underwent reconstruction with a jejunum or latissimus dorsi flap during the same period but were excluded from this study to obtain a homogeneous group of patients. There were 18 men and three women. Mean age of patients at time of recurrence was 59 years (range, years). All patients were treated by CTLP as salvage surgery for recurrence of advanced SCC of the hypopharynx (all stage III or IV at initial assessment). Patients were initially treated by CRT in 14 cases or by surgery followed by postoperative RT or CRT in seven cases. Among the seven patients who initially underwent surgery, four underwent a pharyngectomy and three underwent a total laryngopharyngectomy with primary mucosal closure (i.e., total laryngectomy with resection of unilateral pyriform fossa and closure with the remaining mucosa). In these cases, salvage surgery was a circular pharyngectomy. In all cases, reconstruction was carried out with tubed RFFF (Fig. 1). Fifteen patients (71%) presenting with major cervical fibrosis received a prophylactic pedicled pectoralis major muscle flap to cover the anastomosis. Moreover, 14 patients (67%) underwent a unilateral (8 cases) or bilateral (6 cases) neck dissection in the same procedure. Lymph node dissections were discussed case by case, depending on previous neck surgery history and results of imaging and physical examination (neck fibrosis) during multidisciplinary meetings according to our guidelines: In cases with no evidence of neck disease, a bilateral neck dissection of levels II to III was performed. In cases of neck disease, a radical or modified radical dissection of levels II to V was performed, depending on the extent of the disease. In cases of previous neck dissection or severe neck fibrosis and no evidence of neck disease, no neck dissection was performed. The main characteristics of patients included in the study are summarized in Table I. All patients had received preoperative locoregional and general workups to assess the feasibility of salvage surgery and to search for distant metastases. All diagnostic and therapeutic decisions concerning patients in this study were discussed at multidisciplinary meetings attended by teams representing head and neck surgery, medical oncology, radiation therapy, pathology, and radiology. Main Outcome Measures and Statistical Analysis Overall survival (OS) and disease-free survival (DFS) were determined by Kaplan-Meier analysis. Univariate and multivariate analysis was performed to determine the impact on oncologic outcomes (OS and DFS) of the following factors: age, sex, comorbidity (using the Kaplan-Feinstein index), type of initial treatment (CRT or surgery followed by RT or CRT), postoperative complications, free flap failure, postoperative fistula, positive margins, positive lymph nodes, and extracapsular spread (when positive lymph nodes). All factors analyzed are summarized in Table I. Statistical analyses were performed using log-rank tests (univariate analyses) and Cox regression models (multivariate analyses). All statistical tests were performed with the R software program for Windows, with a 5% threshold of significance. RESULTS Morbidity and Postoperative Course Among the 21 patients who underwent surgery, 15 (71%) presented with postoperative local complications: 911

3 TABLE I. Characteristics of the Study Population. No. of patients 21 Male:female ratio 18:3 Mean age (range), yr 59 (45 78) Kaplan-Feinstein index of comorbidity (%) 2 14 (67) >2 7 (33) Initial treatment (%)* CRT 14 (67) Surgery followed by RT or CRT 7 (33) Type of salvage surgery (%) Circular total laryngopharyngectomy 18 (86) Circular pharyngectomy 3 (14) Reconstruction (%) Free flap alone 6 (29) Free flap þ pedicled pectoralis major flap 15 (71) Neck dissection (%) 14 (67) Unilateral 8 Bilateral 6 Postoperative complications (%) Local (flap failure, infection, hematoma, fistula) 15 (71) General 3 (14) Histopathologic results (%) R0 margins 14 (67) R1 margins 7 (33) pn negative 8/14 (57) pn positive 6/14 (43) Postoperative feeding (%) Oral feeding 13 (62) Gastrostomy 6 (28) Oral þ gastrostomy 2 (10) *All patients initially had advanced hypopharyngeal squamous cell carcinoma (stage III IV). These patients had been previously treated by total laryngectomy with primary mucosal closure. CRT ¼ chemoradiotherapy; RT ¼ radiotherapy. free flap failure in six cases, operative site infection in six, hematoma in three, and fistula in 15. Moreover, three patients (14%) had general complications (sepsis in 2 and hyponatremia in 1). One patient died 3 months after surgery of carotid artery rupture due to a fistula. Among the six patients presenting with free flap failure, a second reconstructive procedure was successfully performed in all cases. Reconstruction was carried out with free flap in three patients (another RFFF in 2 and anterolateral thigh flap [ALT] in 1) and with a pedicled musculocutaneous latissimus dorsi flap in three. Mean hospital stay for the entire study population was 40 days (range, days). Functional Recovery At the end of follow-up, 13 patients (62%) were able to resume oral feeding, six (28%) were fed by gastrostomy, and two (10%) were fed by both oral diet and gastrostomy. 912 Fig. 2. Overall survival. Speech rehabilitation with either esophageal speech or tracheoesophageal prosthesis could not be evaluated due to the retrospective design of the study and the high number of patients who had died at the time of data collection. Histopathologic Results Clear margins (R0) could be obtained in 14 patients (67%), and microscopic positive margins (R1) were seen in seven patients (33%). No patient presented with macroscopic incomplete resection (R2). Among the 14 patients who underwent neck dissection, histopathologic examination showed positive nodes in six (43%) and negative nodes in eight (57%). Extracapsular spread was found in four patients. Oncologic Results After a median follow-up of 30 months from the date of salvage surgery (3 56 months), 15 patients (71%) experienced a recurrence: locoregional recurrence in 12 cases (80%) and distant metastases in three (20%). Median time between surgery and recurrence was 8.7 months. Fourteen patients (67%) died after a median of 11.5 months. At the end of follow-up, six patients were alive with no evidence of disease. The 1, 2, and 5-year OS rates were 46%, 40%, and 16%, respectively (Fig. 2). The 1, 2 and 5-year DFS rates were 42%, 30%, and 15%, respectively (Fig. 3). Predictors of Oncologic Outcomes In our series, two variables were found to have a statistically significant negative impact on both OS and DFS, in either univariate or multivariate analysis. These were free flap failure (P [log rank] ¼.014 and P [Cox] ¼.016) (Fig. 4) and positive margins (P [log rank] ¼.001 and P [Cox] ¼.001) (Fig. 5). All statistical results are summarized in Table II.

4 Fig. 3. Disease-free survival. DISCUSSION The management of hypopharyngeal cancer recurrence after RT or CRT remains challenging, and only surgery can be performed with curative intent. An important point is that, in many cases of patients presenting with a large hypopharyngeal recurrence, the tumor is nonresectable or patient comorbidities do not permit surgery or patients refuse surgery. In the end, only a few patients undergo surgery, which accounts for the small number of patients in our series representing the activity of two tertiary referral centers. Moreover, it is important to bear in mind that this kind of surgery entails a high rate of postoperative complications in patients, with a poor prognosis in terms of survival. Patients eligible for surgery must therefore be selected with care. 5,6,10 Fig. 5. Impact of free flap failure on overall survival. In this study, 71% of patients presented with postoperative local complications. Free flap failure was observed in 28% of cases. There are two reasons for this high complication rate. On the one hand, most of the patients presented with poor local conditions, with very severe cervical fibrosis resulting from previous radiation therapy and/or surgery. Indeed, in our series, 67% of patients had been previously treated by CRT as part of organ preservation protocols, and 33% had been treated by surgery followed by RT or CRT. Among these latter patients, three presented with recurrence following a total laryngectomy and therefore required circular pharyngectomy in very scarred tissue. On the other hand, all patients presented with a high rate of comorbidities, TABLE II. Analysis of Parameters That Could Influence Long-term Oncologic Outcomes. Disease-Free Survival Univariate (Log-Rank) Multivariate (Cox) Overall Survival Univariate (Log-Rank) Multivariate (Cox) Age Sex Initial treatment* Comorbidity Positive margins pnþ Extracapsular spread Flap failure Local complications Fistulas Fig. 4. Impact of positive margins on overall survival. *Initial treatment: chemoradiotherapy versus surgery plus postoperative chemoradiotherapy. Comorbidity was analyzed using the Kaplan-Feinstein index. Statistically significant result. 913

5 since all had a Kaplan-Feinstein index greater than or equal to 2. In a previous study in 2008 carried out at the Lacassagne Comprehensive Cancer Center in Nice, France, analyzing the postoperative course of 213 patients treated by free flap reconstruction, we achieved an overall free flap success rate of 93.4% and a local complication rate of 20.9%. 2 Moreover, this series included some patients from the current study. In this previous study, we showed that a high level of comorbidity (P ¼.009), salvage surgery for recurrent cancer (P ¼.03), and hypopharyngeal surgery (P ¼.002) were associated with a higher risk of local complications. Moreover, salvage surgery for recurrent cancer was the only factor correlated with a higher risk of free flap failure (P ¼.0004). 2 Although the impact of RT on free flap failure risk is still contested by some authors, it is now clearly acknowledged that the combination of chemotherapy and RT increases the difficulties of salvage surgery. 3 6 However, it is difficult to compare our results with other series because of the high heterogeneity between series in terms of patient selection criteria, previous treatments (RT, CRT, and/or surgery), means of reconstruction, and the frequent retrospective design of series. In our series, the choice of type of reconstruction was guided by the desire to avoid abdominal surgery in patients who were often in poor general condition and had a poor prognosis (5-year OS of 16% in our series). Surgical decisions were made case by case based on discussions during multidisciplinary meetings (tumor resectability, neck fibrosis, presence of massive neck invasion, vascular conditions) and the patient s opinion. However, the high rate of flap failure in this series suggests the need to reconsider the choice of type of reconstruction. Many reconstruction techniques, from regional pedicled flaps to free flaps, have already been described. 5 Regarding pedicled musculocutaneous flaps, whether pectoralis major or latissimus dorsi flaps, their bulkiness often limits tubing and makes them somewhat malleable, particularly in cases of postradiation fibrosis. However, horseshoe-shaped (or U-shaped) pectoralis major flaps (i.e., directly sutured to the prevertebral fascia) have been shown to provide good results. 11 Free flaps can be fasciocutaneous (e.g., RFFF), musculocutaneous (e.g., ALT), or digestive. The RFFF has the advantage of allowing easy harvesting, and its excellent plasticity facilitates shaping. It is also known to be generally highly reliable with a very low failure rate. 12,13 This low failure rate was not verified in our CTLP series. In contrast, the RFFF has the disadvantage of producing a relatively large number of fistulas during reconstruction after CTLP. However, the use of a prophylactic pedicled pectoralis major muscle flap to cover the anastomoses between flap and mucosa overcomes this problem and permits rapid healing of these fistulas. Moreover, the pectoralis major flap helps protect vascular anastomoses and cervical vessels from saliva that can sometimes trigger bleeding or carotid rupture. The RFFF is also claimed to produce stricture at the junction between the flap and the esophagus, although 914 very few studies have reported the postoperative outcomes of this flap as a post-ctlp reconstruction. 5 In our series, it was difficult to correctly assess the quality of swallowing and speech of patients owing to the retrospective nature of our data collection and the large number of patients who had died before data collection. Nevertheless, more than 70% of our patients were able to resume oral feeding. The ALT is associated with low surgical morbidity and relatively good functional outcomes. The proponents of the ALT flap argue that, as the donor site is closed primarily, there are fewer soft-tissue donor site complications as compared with the RFFF. Furthermore, the more robust fascia encountered when raising the ALT flap provides a vascularized second layer over the mucosal anastomosis, which likely contributes to the lower reported rates of fistula and stricture as compared with the RFFF. 5 Lastly, the digestive free flaps (jejunum and particularly gastro-omental) seem to yield the best results, including, above all, in salvage surgery, as they allow mucosa-to-mucosa sutures. Many studies have reported excellent results with these flaps. 8,14,15 However, it must be borne in mind that this surgery should be limited to patients in good general condition or, at all events, able to withstand abdominal surgery. 16 It is difficult therefore to ascertain how patients were selected in series publishing excellent postoperative outcomes after reconstruction with digestive flaps and whether or not there was a selection bias due to exclusion of patients in poor condition and those having undergone other reconstructive procedures. 4,5,8,14,17 Finally, there are other techniques, such as gastric transposition (gastric pull-up), that produce good functional results. However, this is also a very invasive procedure with a high morbidity rate and significant perioperative mortality. Consequently, it has limited indications and should be reserved for patients in very good condition. 10,15,18 Although there is abundant research assessing the different methods of reconstruction after circular laryngopharyngectomy, relatively few recent studies have focused on the oncologic results in these patients, especially when treated with salvage surgery. 4,5,7,17 In patients receiving total laryngectomy with primary mucosal closure, 5-year OS ranges from 29% to 66%, depending on the study. 6 In hypopharyngeal advanced cancers, including primary and salvage treatment, 5- year OS ranges from 17% to around 50%. 5,8,17,19 In our series, the 5-year OS was 16%, which is relatively low compared with rates in previously published studies. However, it is important to note that all our cases presented with a recurrence of advanced hypopharyngeal cancer after optimal initial treatment. Moreover, all presented with a poor local condition with significant neck fibrosis. In our series, the two variables that were found to have a statistically significant negative impact on both OS and DFS were free flap failure and microscopic positive margins. Indeed, in our series, all patients presenting with positive margins died within a year following salvage surgery. These results are consistent

6 with previous reports and emphasize the importance of careful selection of patients eligible for this surgery to maximize the chances of achieving clear margins. 6 The influence of flap failure on OS can be explained by the fact that a second reconstruction procedure has a negative impact on the general condition of patients already experiencing multiple comorbidities. Its influence on DFS is more difficult to explain: local inflammation caused by the long period of healing may promote tumor recurrence. 20 However, in our study we found no impact either on OS or on DFS of the presence of salivary fistulas, which also trigger chronic inflammation. In our series, we also observed no other factors influencing survival, including the presence of positive nodes, which in our view is not a contraindication for salvage CTLP. Finally, after a median follow-up of 30 months, six patients (29%) were alive and free of disease. All these patients had clear margins at histopathologic examination of surgical specimens. None had free flap failure. CONCLUSION Salvage surgery remains the only curative option for treatment of recurrent hypopharyngeal SCC. Its goal is to obtain tumor control and satisfactory quality of life. Despite the high rate of postoperative complications in our series, this type of treatment should be offered to patients whenever possible. However, selection of patients eligible for this type of surgery is important since long-term survival is clearly correlated with the presence of clear margins and the absence of free flap failure. Thus, this treatment should not be considered as "palliative" or "debulking" surgery. Indeed, in our series, all patients presenting with positive margins died within the year following salvage surgery. Concerning the method of reconstruction, our results suggest that, in cases of recurrence after treatment with associated CRT or when cervical post-rt fibrosis is very significant (as with patients in our series), it is preferable, if the patient is in good condition, to perform a first-line digestive free flap. ALT flap is also a good and less invasive option, particularly when patients are in poor condition. BIBLIOGRAPHY 1. Lefebvre JL, Lartigau E. Preservation of form and function during management of cancer of the larynx and hypopharynx. World J Surg 2003; 27: Dassonville O, Poissonnet G, Chamorey E, et al. Head and neck reconstruction with free flaps: a report on 213 cases. Eur Arch Otorhinolaryngol 2008;265: Ganly I, Patel S, Matsuo J, et al. Postoperative complications of salvage total laryngectomy. Cancer 2005;103: Miyamoto S, Sakuraba M, Nagamatsu S, Hayashi R. Salvage total pharyngolaryngectomy and free jejunum transfer. Laryngoscope 2011;121: Patel RS, Goldstein DP, Brown D, Irish J, Gullane PJ, Gilbert RW. Circumferential pharyngeal reconstruction: history, critical analysis of techniques, and current therapeutic recommendations. Head Neck 2010;32: Van der Putten L, de Bree R, Kuik DJ, et al. Salvage laryngectomy: oncological and functional outcome. Oral Oncol 2011;47: Withrow KP, Rosenthal EL, Gourin CG, et al. Free tissue transfer to manage salvage laryngectomy defects after organ preservation failure. Laryngoscope 2007;117: Kadota H, Fukushima J, Nakashima T, et al. Comparison of salvage and planned pharyngolaryngectomy with jejunal transfer for hypopharyngeal carcinoma after chemoradiotherapy. Laryngoscope 2010;120: Patel RS, Makitie AA, Goldstein DP, et al. Morbidity and functional outcomes following gastro-omental free flap reconstruction of circumferential pharyngeal defects. Head Neck 2009;31: Clark JR, Gilbert R, Irish J, Brown D, Neligan P, Gullane PJ. Morbidity after flap reconstruction of hypopharyngeal defects. Laryngoscope 2006; 116: Jegoux F, Ferron C, Malard O, Espitalier F, Beauvillain de Montreuil C. Reconstruction of circumferential pharyngolaryngectomy using a horseshoe-shaped pectoralis major myocutaneous flap. J Laryngol Otol 2007; 121: Bozec A, Poissonnet G, Chamorey E, et al. Quality of life after oral and oropharyngeal reconstruction with a radial forearm free flap: prospective study. J Otolaryngol Head Neck Surg 2009;38: Bozec A, Poissonnet G, Chamorey E, et al. Free-flap head and neck reconstruction and quality of life: a 2-year prospective study. Laryngoscope 2008;118: Moradi P, Glass GE, Atherton DD, et al. Reconstruction of pharyngolaryngectomy defects using the jejunal free flap: a 10-year experience from a single reconstructive center. Plast Reconstr Surg 2010;126: Julieron M, Germain MA, Schwaab G, et al. Reconstruction with free jejunal autograft after circumferential pharyngolaryngectomy: eighty-three cases. Ann Otol Rhinol Laryngol 1998;107: Dubsky PC, Stift A, Rath T, Kornfehl J. Salvage surgery for recurrent carcinoma of the hypopharynx and reconstruction using jejunal free tissue transfer and pectoralis major muscle pedicled flap. Arch Otolaryngol Head Neck Surg 2007;133: Clark JR, de Almeida J, Gilbert R, et al. Primary and salvage (hypo) pharyngectomy: Analysis and outcome. Head Neck 2006;28: Mariette C, Fabre S, Balon JM, Patenotre P, Chevalier D, Triboulet JP. Reconstruction after total circular pharyngo-laryngectomy: comparison between gastric interposition and free jejunal flap. Ann Chir 2002;127: Kim S, Wu HG, Heo DS, Kim KH, Sung MW, Park CI. Advanced hypopharyngeal carcinoma treatment results according to treatment modalities. Head Neck 2001;23: Maher SG, Reynolds JV. Basic concepts of inflammation and its role in carcinogenesis. Recent Results Cancer Res 2011;185: