Locoregional recurrences are the most frequent

ORIGINAL ARTICLE SECOND SALVAGE SURGERY FOR RE-RECURRENT ORAL CAVITY AND OROPHARYNX CARCINOMA Ivan Marcelo Gonçalves Agra, MD, PhD, 1 João Gonçalves Filho, MD, PhD, 2 Everton Pontes Martins, MD, PhD, 2 Luiz Paulo Kowalski, MD, PhD 2 1 Head and Neck Surgery Department, Hospital Aristides Maltez, Salvador, Brazil 2 Head and Neck Surgery and Otorhinolaryngology Department, Hospital AC Camargo, São Paulo, Brazil. E-mail: jgon.13@terra.com.br Accepted 14 September 2009 Published online 26 February 2010 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/hed.21298 Abstract: Background. Salvage surgery is considered the best treatment approach for patients with recurrent oral carcinoma. Unfortunately, 50% to 60% of the patients who undergo salvage surgery will develop further locoregional recurrence, and they are not usually considered for further treatment. Our aim in this study was to report our experience with a second salvage surgery for selected patients with re-recurrent oral cavity and oropharyngeal squamous cell carcinoma (SCC). Methods. Forty-one patients underwent a second salvage surgical procedure, with curative intention for re-recurrent oral cancer. The surgical treatment used was wide local resection in 34 cases, neck dissection in 9 cases, and isolated neck dissection in 7 cases. Results. Cancer-specific survival (CSS) rate in 3 years was at 20%. Patients with re-recurrence in <6 months presented 3-year CSS null, whereas patients with re-recurrence after 6 months presented 3-year CSS of 32.3% (p ¼.007). Conclusion. Second salvage surgery can be considered a potentially curative therapeutic approach for a selected group of patients with re-recurrent oral SCC. The disease-free interval was the main clinical factor associated with the prognosis. VC 2010 Wiley Periodicals, Inc. Head Neck 32: 997 1002, 2010 Keywords: salvage surgery; oral cavity; oropharynx; prognostic factors; recurrences Correspondence to: J. G. Filho VC 2010 Wiley Periodicals, Inc. Locoregional recurrences are the most frequent cause of treatment failure of patients with squamous cell carcinoma (SCC) of the oral cavity and oropharynx. The rates of recurrence vary according to the TNM stage and initial treatment, and can be found in >50% of the patients with tumors diagnosed at advanced clinical stages (CS III/IV). 1,2 The best treatment approach for these patients is salvage surgery for those with good clinical performance status and resectable tumors. 3,4 Approximately 60% of the patients who undergo salvage surgery will present with a new locoregional recurrence. Isolated distant metastases continue to be an infrequent event, occurring in <20% of these patients. 4 Second salvage surgery is usually not indicated in most institutions, and the patients undergo palliative treatments or the best supportive care. However, Cherian et al 5 used a second salvage procedure as a therapeutic option for a very select group of patients seen with further locoregional recurrence. The challenge was to identify the patients for whom the cost and morbidity of these procedures would justify a new intervention. Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010 997

The object of this study was to analyze our experience with a second salvage surgery for recurrent oral cavity and oropharyngeal SCC, the results of this therapeutic strategy, and the factors associated with the prognosis. MATERIALS AND METHODS Table 1. Characteristics of the patients with re-recurrent oral cancer. Variable No. (%) Age, y Median 55 Range 26-80 Sex Male 31 (75.6) Female 10 (24.4) Site of tumor Lip 7 (17.1) Floor of mouth 6 (14.5) Tongue 10 (24.5) Buccal mucosa 2 (4.9) Gingiva 2 (4.9) Retromolar trigone 2 (4.9) Soft palate 4 (9.8) Base of tongue 2 (4.9) Tonsil 6 (14.5) Clinical stage rcs I-II 7 (17.1) rcs III-IV 34 (82.9) Type of recurrence Local 34 (82.9) Regional 7 (17.1) Disease-free interval <6 mo 14 (34.1) >6 mo 27 (65.9) Abbreviation: rcs, clinical staging of recurrence. Between 1990 and 2004, 246 patients underwent salvage surgery for recurrent SCC of the oral cavity and oropharynx at the Hospital AC Camargo, in São Paulo, Brazil. 4 In all, 135 patients (54.9%) had further recurrence during the follow-up period. In 119 cases, there was isolated locoregional recurrence (88.1%). Forty-one patients had resectable tumors, had no distant metastasis, presented good health conditions, and accepted a second salvage surgery. In this group, histopathologic studies of the margins after first salvage surgery showed free margins in 33 patients (80.5%), close margins in 3 patients (7.3%), and positive margins in 5 patients (12.2%). Adjuvant radiation therapy had been performed in 13 patients (all of the patients with positive or close margins, and 5 patients with free margins, but with the disease at an advanced stage). The patients ages ranged from 26 to 80 years, with a median of 55 years; 31 patients were men and 10 were women. The site of the primary tumor was the oral cavity in 29 patients (70.7%) and the oropharynx in 12 patients (29.3%). The type of recurrence was local in 34 patients (82.9%) and regional in 7 patients (17.1%). The clinical staging of second recurrence was classified as initial (rcs I/II) in 7 patients (17.1%) and advanced (rcs III/IV) in 34 patients (82.1%) (Table 1). Sixteen of the patients were initially treated with surgery alone, 15 patients with radiotherapy alone, and 10 patients with surgery and adjuvant radiotherapy. For the 7 patients that had recurrence in the neck, salvage surgery was radical neck dissection in all cases. One patient underwent craniofacial resection (recurrent retromolar trigone carcinoma in the infratemporal fossa). In 16 patients, surgery consisted of wide resection of the tumor with or without mandibulotomy. In 12 patients, a wide resection was performed including a mandibulectomy. Five patients presenting with re-recurrent oropharynx SCC underwent total laryngectomy or pharyngolaryngectomy because there was larynx invasion. Nine patients underwent en bloc surgeries with a neck dissection. The type of reconstruction used was primary closure in 6 patients, local flaps in 21 patients, Backanjian s flap in 2 patients, pectoralis major myocutaneous flap in 10 patients, and microvascular free flaps in 2 patients. Hospitalization time ranged from 2 to 30 days (median, 5 days). Thirty-eight patients (92.6%) had previously received a radical dose of radiotherapy. Three patients were treated with this therapeutic modality after the resalvage surgery. In 5 patients, reirradiation was performed: in 3 patients by means of brachytherapy and in 2 patients with teleradiotherapy. Surgery was considered with adequate macroscopic margins in 39 patients (95.2%). In 2 patients, complete resection of the tumor was not possible because in 1 of these patients the residual tumor was in the field of node dissection and, in the other, at the skull base. However, the surgical margins (microscopic) were free in only 25 patients (61%), close in 6 patients (14.6%), and there were positive margins in 10 patients (24.4%). Eight patients who 998 Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010

Table 2. Characteristics of 11 patients who did not develop new recurrence. Patient no./age, y/sex Tumor site First treatment rcs Second salvage treatment Follow-up, mo 1/31/M Tonsil S IV S 13.3 2/49/M Tongue S III S 0.5 2/49/F Tongue S IV S 68.0 4/53/M Soft palate RT IV S 11.6 5/56/F Soft palate RT III S 106.0 6/56/M Soft palate S IV S 1.0 7/85/M Retromolar trigone S II S 5.3 8/62/F Lip S IV S 6.7 9/62/M Tonsil RT IV S 11.6 10/66/M Lip S I S 4.8 11/72/F Soft palate RT IV S 12.1 Abbreviations: S, surgery; RT, radiotherapy; rcs, clinical staging of recurrence. had positive or close margins after second salvage surgery received adjuvant treatment with radiation therapy (in 3 patients with interstitial high-dose-rate brachytherapy and in 5 patients with external-beam radiation therapy). The SPSS program (version 10.0) 6 was used for the statistical analysis. The survival analysis was done using the Kaplan Meier 7 method and the log-rank test. The disease-free interval was defined as the time elapsed (in months) from the date of the previous salvage surgery to the diagnosis of the new recurrence. The cancer-specific survival (CSS) was defined as the interval between the date of second salvage surgery and the last objective information of follow-up or cancer-specific death. Statistical significance was set for values of p <.05. RESULTS Follow-up ranged from 0.5 to 122.9 months, with a median of 12.1 months. Three patients were lost to follow-up (7.3%). At the end of this study, 11 patients (26.8%) did not present a new recurrence of the disease, after a median followup period of 11.6 months (range, 0.5 to 106 months) (Table 2). However, 30 patients (73.2%) had further recurrence. Twenty-four patients had locoregional recurrence (58.5%), whereas 4 patients had isolated distant metastases (9.7%). Two patients had concomitant local recurrence and distant metastasis. Eight patients with locoregional recurrences underwent a third salvage surgery. Among the other 22 patients, 14 had only supportive care, 2 were treated with reirradiation, and 6 were treated with systemic chemotherapy. There were no surgical complications in 32 patients (78%). Nine patients had surgical complications (22%). In 6 patients, there was operative wound infection, orocutaneous fistula occurred in 6 patients, and necrosis of the reconstruction flaps occurred in 4 patients. One patient had carotid blowout that was treated with ligature. One patient died after surgery as a result of stroke (postoperative mortality of 2.4%). The median recurrence-free survival rate was of 5.76 months. Locoregional control was obtained in 15 patients (36.6%) during followup. Overall survival in 3 years was of 18%. CSS in 3 years was of 20% (see Figure 1). The variables site of tumor and type of recurrence were not associated with statistically significant differences in the rate of CSS. However, the female sex was associated with a significantly better rate of CSS (Table 3). The disease-free interval, defined as the time interval between the first salvage surgery and the date of the second recurrence, ranged from FIGURE 1. Cancer-specific survival. Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010 999

Table 3. Cancer-specific survival in accord with the variables. Variable Survival at 3 y, % p value Sex.043 Male (n ¼ 31) 9.7 Female (n ¼ 10) 45.7 Tumor site.342 Oral cavity (n ¼ 29) 13.1 Oropharynx (n ¼ 12) 40.9 Clinical stage.312 rcs I-II (n ¼ 7) 50.0 rcs III-IV (n ¼ 34) 12.5 Recurrence.313 Local (n ¼ 33) 21.2 Regional (n ¼ 8) 14.3 Surgical margins.403 Negative (n ¼ 25) 25.8 Close/positive (n ¼ 16) 9.3 Disease-free interval.007 <6 mo(n ¼ 14) 0.0 >6 mo(n ¼ 27) 33.2 Abbreviation: rcs, clinical staging of recurrence. FIGURE 2. Cancer-specific survival in accord with disease-free interval after the first salvage surgery.! interval until 6 months; n interval after 6 months; p ¼.007. FIGURE 3. Cancer-specific survival in accord with clinical staging of second recurrence.! rcs I/II; n rcs III/IV; p ¼.312. rcs, clinical staging of recurrence. 1.5 to 39.7 months, with a median of 7.7 months. The casuistic population was divided into patients who presented recurrence until 6 months after the first salvage surgery (14 patients) and patients who presented recurrence after 6 months (27 patients). The CSS in 3 years was null for patients with early recurrence (until 6 months), and 33.2% for those with late recurrence (after 6 months) (p ¼.007) (see Figure 2). Patients with recurrences with rcs I/II presented CSS in 3 years of 50.0%, whereas patients with rcs III/IV attained survival of 12.5% (p ¼.312) (Table 3). Patients with free margins in surgical specimen had a better prognosis than that of patients with close or positive margins (3-year CSS, 25.8% and 9.3%, respectively). However, this difference was not statistically significant (p ¼.403) (see Figure 3). DISCUSSION Approximately 50% to 60% of the patients who underwent salvage surgery for SCC of the oral cavity and oropharynx experienced further locoregional recurrences. Distant metastases continued to be rather uncommon even among these patients, and represented <20% of the recurrence sites. A significant number of patients with locoregional recurrences had good clinical conditions, had no significant comorbidities, and were motivated to undergo a new salvage treatment. Salvage surgery can be the only curative alternative for selected patients. The results of chemotherapy alone and reirradiation associated with chemotherapy in such patients are very poor, and the expected rate of survival is just a few months. 8,9 The CSS results obtained in this series can be considered modest, although salvage surgery is the only potentially curative alternative for these patients. One current approach for treating locally advanced or recurrent tumors of the head and neck is intraoperative radiation therapy (IORT), 10 mainly because salvage therapy in areas were prior radiation limits and external beam radiation therapy. 11 However, some authors have reported that IORT alone is not sufficient for long-term control in patients with recurrent head and neck cancer. 12,13 Recently, Chen et al 14 reported 51% of locoregional control at 3 years after salvage surgery and IORT in a series of 137 patients treated in a single institution. The authors considered that surgical resection plus IORT is an effective treatment 1000 Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010

modality for selected patients with recurrent head and neck cancer, with good rates of locoregional control and with acceptable morbidity. In our series, IORT was not used, but the use of intraoperative interstitial implantation for interstitial high-dose-rate brachytherapy was an option in 30% of the patients (3/10) with positive margins in this series. A previous study from our institution showed that 52% of patients with neck recurrence experienced neck control after surgery and brachytherapy. 15,16 Cherian et al 5 presented their experience with salvage surgery in 78 patients with recurrent SCC of the buccal mucosa after radiotherapy. Thirty-one patients (39.7%) had further recurrence, and in 13 patients with locoregional recurrence a second salvage surgery was performed. Five of the 13 patients had no recurrence up to the end of follow-up. We were not able to find further information about resalvage procedures in the literature. The clinical staging of the second recurrence was not a variable that influenced the prognosis in this series. We believe that the staging presented no statistical difference attributed to the bad prognosis that the patients already presented as a result of the aggressiveness of the disease. This difference will probably be confirmed only with a larger sample of patients. Disease-free interval seems to be a variable to be considered in the therapeutic decision; patients with recurrence after 6 months have a better prognosis than those patients who have early recurrences. These results strongly suggest that patients with early recurrences (<6 months) should be referred to palliative treatment or best supportive care. Free surgical margins were associated with a better prognosis, and, despite no statistically significant differences resulting from the small sample size in the studied patients, it must be considered an important goal of the salvage surgery thus, the importance of adequate surgical planning, based on the careful interpretation of CT scans, nuclear magnetic resonance, or ideally PET-CT. Careful examination in the operating room under general anesthesia, by an experienced surgeon, with focus on the limits of the tumor, which is frequently difficult to define, seems to be a fundamental step toward the success of treatment by means of surgical resection with free margins. Frozen-section assessment of surgical margin should be routinely used in these cases. Biological markers could be useful for therapeutic decision. Epidermal growth factor receptor (EGFR) expression by means of immunohistochemical examination could be an important tool for defining patients with better prognosis, as shown in a previous study, 17 and possibly could be used to differentiate patients with the best prognosis (those not expressing this protein). The selection of patients for second salvage surgery must be very careful. In our opinion, the recommendation for these procedures must be made individually, case by case. The patients should be motivated and conscious of the risks of the procedure. They must have good conditions of health, resectable tumors allowing surgery with curative intention, and the possibility of adequate rehabilitation. It is also very important that the disease-free interval should be >6 months from the first salvage surgery. In conclusion, second salvage surgery in SCC of the oral cavity and oropharynx is a potentially curative therapeutic approach for a very select group of patients who have previously undergone salvage surgery. The disease-free interval was the main clinical factor associated with the prognosis in this series; patients who had recurrence in a period >6 months after the first salvage surgery had a better prognosis. REFERENCES 1. Kowalski LP, Bagietto R, Lara JRL, Santos RL, Silva JF Jr, Magrin J. Prognostic significance of the distribution of neck node metastasis from oral carcinoma. Head Neck 2000;22:207 214. 2. Wong LY, Wei WI, Lam LK, Yuen APW. Salvage of recurrent head and neck squamous cell carcinoma after primary curative surgery. Head Neck 2003;25:953 957. 3. Goodwin W Jr. Salvage surgery for patients with recurrent squamous cell carcinoma of the upper aerodigestive tract: when do the ends justify the means? Laryngoscope 2000;110 (Suppl 93):1 18. 4. Agra IM, Carvalho AL, Kowalski LP, et al. Prognostic factors in salvage surgery for recurrent oral and oropharyngeal cancer. Head Neck 2006;28:107 113. 5. Cherian T, Sebastian P, Ahamed MI, et al. Evaluation of salvage surgery in heavily irradiated cancer of the buccal mucosa. Cancer 1991;68:295 299. 6. Statistical Package for the Social Sciences (SPSS). Statistical data analysis software for Windows. Chicago: SPSS Inc.; 2000. 7. Kaplan EL, Meier P. Nonparametric simulation from incomplete observations. J Am Static Assoc 1958;53: 457 481. 8. Wong SJ, Machtay M, Li Y. Locally recurrent, previously irradiated head and neck cancer: concurrent re-irradiation and chemotherapy, or chemotherapy alone? J Clin Oncol 2006;24:2653 2658. 9. Colevas AD. Chemotherapy options for patients with metastatic or recurrent squamous cell carcinoma of the head and neck. J Clin Oncol 2006;24:2644 2652. Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010 1001

10. Marucci L, Pichi B, Laccarino G, Ruscito P, Spriano G, Arcangeli G. Intraoperative radiation therapy as an early boost in locally advanced head and neck cancer: preliminary results of a feasibility study. Head Neck 2008;30:701 708. 11. Most MD, Allori AC, Hu K, et al. Feasibility of flap reconstruction in conjunction with intraoperative radiation therapy for advanced and recurrent head and neck cancer. Laryngoscope 2008;118:69 74. 12. Nag S, Schuller DE, Martinez-Monge R, Rodriguez-Villalba S, Grecula J, Bauer C. Intraoperative electron beam radiotherapy for previously irradiated advanced head and neck malignancies. Int J Radiat Oncol Biol Phys 1998;42:1085 1089. 13. Schleicher UM, Phonias C, Spaeth J, Schlöndorff G, Ammon J, Andreopoulos D. Intraoperative radiotherapy for pre-irradiated head and neck cancer. Radiother Oncol 2001;58:77 81. 14. Chen AM, Bucci MK, Singer MI, et al. Intraoperative radiation therapy for recurrent head-and-neck cancer: the UCSF experience. Int J Radiat Oncol Biol Phys 2007;67:122 129. 15. Pellizzon AC, dos Santos Novaes PE, Conte Maia MA, et al. Interstitial high-dose-rate brachytherapy combined with cervical dissection on head and neck cancer. Head Neck 2005;27:1035 1041. 16. Pellizzon AC, Salvajoli JV, Kowalski LP, Carvalho AL. Salvage for cervical recurrences of head and neck cancer with dissection and interstitial high dose rate brachytherapy. Radiat Oncol 2006;1:27 (abstract). 17. Agra IM, Carvalho AL, Pinto CAL, et al. Biological markers and prognosis in recurrent oral cancer submitted to salvage surgery. Arch Otolaryngol Head Neck Surg 2008;134:743 749. 1002 Second Salvage Surgery HEAD & NECK DOI 10.1002/hed August 2010