ORIGINAL ARTICLE
Clinical outcome and prognostic factors after salvage surgery for isolated regional squamous cell carcinoma recurrences Eun-Jae Chung, MD, PhD,1 Sang-Hyo Lee, MD,1 So-Hye Baek, MD,1 Woo-Jin Bae, MD,1 Yong-Joon Chang, MD,2 Young-Soo Rho, MD, PhD3* 1
Department of Otorhinolaryngology–Head and Neck Surgery, Ilsong Memorial Institute of Head and Neck Cancer, Hallym University, College of Medicine, Seoul, Korea, Department of Plastic and Reconstructive Surgery, Ilsong Memorial Institute of Head and Neck Cancer, Hallym University, College of Medicine, Seoul, Korea, 3Department of Otorhinolaryngology–Head and Neck Surgery, Ewha Womans University, College of Medicine, Seoul, Korea.
2
Accepted 11 June 2014 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23799
ABSTRACT: Background. The purpose of this study was to evaluate the outcome and predictive factors for salvage surgery of isolated regional recurrences of head and neck squamous cell carcinoma. Methods. A retrospective study was conducted with 55 patients who were treated with surgery-based treatment. Results. The 5-year overall survival (OS) and disease-free survival (DFS) rates were 61.8% and 60%, respectively. Extracapsular spread (ECS) was an independent factor associated with worse diseasespecific survival. The patients who had advanced N classification, ECS, and in-field recurrence had a significantly worse OS rate,
whereas those with an initial DFS time of 6 months or more experienced better outcomes. Conclusion. Salvage surgery for isolated regional recurrence resulted in an acceptable oncologic outcome and mortality. Successful surgical salvage is most probable in late recurrence (6 months) patients with recurrent N1 stage tumors (no evidence of ECS) outside of the previous treatment field. C 2014 Wiley Periodicals, Inc. Head Neck 00: 000–000, 2014 V
INTRODUCTION
14-2-21). The records of 328 patients who received surgical treatment for recurrences at the Ilsong Memorial Institute Head and Neck Cancer, Hallym University College of Medicine, from 1993 through 2012 were retrospectively identified and reviewed. Patients with head and neck tumors other than squamous cell carcinoma, stomal recurrence, who underwent palliative surgery, or who presented with second primary or with distant metastasis were excluded. Consequently, 203 patients who underwent salvage surgery for recurrent HNSCC were enrolled. Of those patients, 55 had regional recurrence, which included the occurrence of lymph node metastasis; 112 had local recurrence, defined as recurrence in the primary site, and 36 had locoregional recurrence, which consisted of both local and regional recurrence. Thus, a total of 55 isolated regional recurrences were retrospectively analyzed in this study. The study group was composed of 43 men and 12 women (mean age, 58.9 years; range, 39–85 years). The follow-up period ranged from 1 month to 201 months (mean, 21.7 months). The original primary tumor site was the oral cavity in 22 patients (40%), the larynx in 13 patients (23.6%), the hypopharynx in 7 patients (12.7%), the oropharynx in 8 patients (14.5%), the nasopharynx in 3 patients (5.5%), and the nasal cavity/maxillary sinus in 2 patients (3.6%). Previous treatment consisted of surgery alone in 16 patients, surgery and radiotherapy with or without chemotherapy in 25 patients, and curative radiotherapy with or without chemotherapy in 14 patients.
Management of recurrence after prior curative treatment is a challenging clinical situation. Generally, salvage cure rates of neck recurrences are thought to be higher than those for local recurrences, but survival rates are not significantly different.1 Salvage neck dissection offers the best curative chance for patients with resectable neck recurrences. However, surgical salvage is often limited by extensive tumor involvement, difficult resection and reconstruction, and the poor general health of patients at the time of recurrence. There is still a lack of uniformity concerning factors that predict the suitability of salvage treatment for isolated neck recurrences. Therefore, it is important to determine which patients would most likely benefit from surgical salvage. The purpose of this study was to evaluate the oncologic outcome and predictive factors for successful curative salvage surgery after isolated nodal recurrence in head and neck squamous cell carcinoma (HNSCC).
MATERIALS AND METHODS Patient population This study was reviewed and approved by the Hallym University Institutional Review Board (IRB number:
*Corresponding author: Y.-S. Rho, Department of Otorhinolaryngology–Head and Neck Surgery, Ewha Woman’s University, College of Medicine, 911-1 Mok-Dong, Yang Cheon-Gu, Seoul, Korea 134-701. E-mail: [email protected]
KEY WORDS: head and neck cancer, recurrence, surgery, prognosis, mortality
HEAD & NECK—DOI 10.1002/HED
MONTH 2014
1
CHUNG ET AL.
TABLE 1. Pathologic N classifications for the study patients. No. of patients (%) by N classification
Oral cavity Oropharynx Nasopharynx Hypopharynx Larynx Nasal cavity/maxillary sinus Total
N1
N2a
N2b
N2c
N3
Total
4 2 2 1 1 0 10
6 0 0 1 2 2 11
6 2 1 3 6 0 18
3 1 0 0 2 0 6
3 3 0 2 2 0 10
22 8 3 7 13 2 55
The recurrent pathologic N classifications (rpN) are summarized in Table 1. The tumors were classified as rpN1 in 10 patients (18.2%), rpN2a in 11 patients (20%), rpN2b in 18 patients (32.7%), rpN2c in 6 patients (10.9%), and rpN3 in 10 patients (18.2%). Postoperative adjuvant treatment (after salvage surgery) was conducted after multidisciplinary discussion. Surgical treatment was followed by postoperative radiotherapy in 5 patients (9.1%), and chemoradiation in 19 patients (34.5%); among these patients, 16 (66.7%) had previously received radiation. Thirty-one patients (56.4%) were treated with surgery alone. The type of neck dissection depended on the extent of disease and the findings at the time of surgery. Eight patients underwent a bilateral neck dissection (3 therapeutic neck dissections and 5 elective neck dissections). Contralateral elective neck dissections were considered for patients who were node positive at initial treatment, presented with advanced rN classification, and developed recurrence within 1 year of initial treatment.2 Twenty-five patients underwent a radical or modified radical neck dissection. Extended radical neck dissections were performed for 26 patients (extended radical neck dissection was defined as the neck dissection that removes additional lymph node groups or nonlymphatic structures, not ordinarily encompassed by radical neck dissection, and that should be done for the removing of metastatic structure not for the approach to primary site). Carotid artery (n 5 17; carotid subadventitial dissection in 13 patients and carotid resection with/without reconstruction in 4 patients) was the most common structure sacrificed during extended radical neck dissection. Nerves were the next most common (n 5 16; including the hypoglossal, vagus, superior laryngeal, lingual, and phrenic nerve), followed by muscle (n 5 9; including the prevertebral, paraspinal, mylohyoid, stylohyoid, and digastric muscles) and lymph node (n 5 8; retropharyngeal, parapharyngeal, and level VII node; Table 2). Selective neck dissections were performed for 4 patients.
Determination of the optimal cut-point for defining early versus late recurrence The disease-free time was defined as the elapsed time from the first day after the end of curative treatment to the time when recurrence was confirmed. Salvage time was defined as the time after completion of the salvage treatment until recurrence, metastasis, or last follow-up. The 5-year disease-free survival (DFS) and overall 2
HEAD & NECK—DOI 10.1002/HED
MONTH 2014
survival (OS) rates were calculated by using Kaplan– Meier analysis after primary definitive treatment. The optimal cut-point for defining early versus late initial relapse (ie, before salvage) was chosen on the basis of 5year DFS and OS.3
Statistical analysis The OS and DFS rates were determined using the Kaplan–Meier actuarial life-table method with the logrank test for statistical comparison. We used Cox proportional hazard models for the multivariate analysis of survival. A p value < .05 was considered significant.
RESULTS Optimal cutoff value for early versus late recurrence Table 3 summarizes the 5-year DFS and OS rates across different cut-points for initial tumor recurrence before salvage (expressed as disease-free time). The cutoff point of 6 months was chosen because it yielded the most significant difference for the DFS and OS rates.
Surgical outcomes The surgical salvage rate (33 of 55 patients) was 60% for isolated regional recurrence. Recurrent disease developed in 22 patients (40%), with local recurrence in TABLE 2. Type of salvage neck dissection for the study patients. Salvage neck dissection
Selective neck dissection Modified radical neck dissection/radical neck dissection Extended radical neck dissection Lymph node Level VI/level VII Retropharyngeal/parapharyngeal Carotid artery Nerve Hypoglossal Vagus Superior laryngeal Lingual Phrenic Skin Glandular structure Total
No. of patients
4 25 26 8 6 5 17 16 9 6 2 3 3 5 2 55
SALVAGE
TREATMENT FOR ISOLATED REGIONAL RECURRENCES
TABLE 3. Determination of the optimal cut-point for defining early versus late recurrence before salvage surgery. 5-y DFS
5-y OS
Disease-free time, mo
< (%)
(%)
p value
< (%)
(%)
p value
4 5 6 7 8 9 10 11 12 13 14 15 16
72.9 50 52.4 54.5 59.3 58.6 58.8 57.5 57.1 57.1 58.1 58.1 58.1
56.8 64.1 64.7 63.6 60.7 61.5 61.9 66.7 69.2 69.2 66.7 66.7 66.7
.792 .088 .02* .094 .248 .229 .167 .157 .149 .149 .225 .225 .225
45.5 50 52.4 54.5 55.6 51.7 55.9 57.5 57.1 57.1 58.1 58.1 58.1
65.9 66.7 67.6 66.7 67.9 73.1 71.4 73.3 76.9 76.9 75 75 75
.027* .043* .007* .05* .05* .037* .076 .108 .084 .084 .122 .122 .122
Abbreviations: DFS, disease-free survival; OS, overall survival. The figures in bold and * values indicate statistical significance.
TABLE 4. Clinicopathological factors influencing disease-free survival.
Parameter
Previous neck treatment No Yes Initial T classification T1–2 T3–4 Initial N classification N0–1 N2–3 Recurrent N classification N0–1 N2–3 Time to recurrence
Clinical outcome and prognostic factors after salvage surgery for isolated regional squamous cell carcinoma recurrences Eun-Jae Chung, MD, PhD,1 Sang-Hyo Lee, MD,1 So-Hye Baek, MD,1 Woo-Jin Bae, MD,1 Yong-Joon Chang, MD,2 Young-Soo Rho, MD, PhD3* 1
Department of Otorhinolaryngology–Head and Neck Surgery, Ilsong Memorial Institute of Head and Neck Cancer, Hallym University, College of Medicine, Seoul, Korea, Department of Plastic and Reconstructive Surgery, Ilsong Memorial Institute of Head and Neck Cancer, Hallym University, College of Medicine, Seoul, Korea, 3Department of Otorhinolaryngology–Head and Neck Surgery, Ewha Womans University, College of Medicine, Seoul, Korea.
2
Accepted 11 June 2014 Published online 00 Month 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/hed.23799
ABSTRACT: Background. The purpose of this study was to evaluate the outcome and predictive factors for salvage surgery of isolated regional recurrences of head and neck squamous cell carcinoma. Methods. A retrospective study was conducted with 55 patients who were treated with surgery-based treatment. Results. The 5-year overall survival (OS) and disease-free survival (DFS) rates were 61.8% and 60%, respectively. Extracapsular spread (ECS) was an independent factor associated with worse diseasespecific survival. The patients who had advanced N classification, ECS, and in-field recurrence had a significantly worse OS rate,
whereas those with an initial DFS time of 6 months or more experienced better outcomes. Conclusion. Salvage surgery for isolated regional recurrence resulted in an acceptable oncologic outcome and mortality. Successful surgical salvage is most probable in late recurrence (6 months) patients with recurrent N1 stage tumors (no evidence of ECS) outside of the previous treatment field. C 2014 Wiley Periodicals, Inc. Head Neck 00: 000–000, 2014 V
INTRODUCTION
14-2-21). The records of 328 patients who received surgical treatment for recurrences at the Ilsong Memorial Institute Head and Neck Cancer, Hallym University College of Medicine, from 1993 through 2012 were retrospectively identified and reviewed. Patients with head and neck tumors other than squamous cell carcinoma, stomal recurrence, who underwent palliative surgery, or who presented with second primary or with distant metastasis were excluded. Consequently, 203 patients who underwent salvage surgery for recurrent HNSCC were enrolled. Of those patients, 55 had regional recurrence, which included the occurrence of lymph node metastasis; 112 had local recurrence, defined as recurrence in the primary site, and 36 had locoregional recurrence, which consisted of both local and regional recurrence. Thus, a total of 55 isolated regional recurrences were retrospectively analyzed in this study. The study group was composed of 43 men and 12 women (mean age, 58.9 years; range, 39–85 years). The follow-up period ranged from 1 month to 201 months (mean, 21.7 months). The original primary tumor site was the oral cavity in 22 patients (40%), the larynx in 13 patients (23.6%), the hypopharynx in 7 patients (12.7%), the oropharynx in 8 patients (14.5%), the nasopharynx in 3 patients (5.5%), and the nasal cavity/maxillary sinus in 2 patients (3.6%). Previous treatment consisted of surgery alone in 16 patients, surgery and radiotherapy with or without chemotherapy in 25 patients, and curative radiotherapy with or without chemotherapy in 14 patients.
Management of recurrence after prior curative treatment is a challenging clinical situation. Generally, salvage cure rates of neck recurrences are thought to be higher than those for local recurrences, but survival rates are not significantly different.1 Salvage neck dissection offers the best curative chance for patients with resectable neck recurrences. However, surgical salvage is often limited by extensive tumor involvement, difficult resection and reconstruction, and the poor general health of patients at the time of recurrence. There is still a lack of uniformity concerning factors that predict the suitability of salvage treatment for isolated neck recurrences. Therefore, it is important to determine which patients would most likely benefit from surgical salvage. The purpose of this study was to evaluate the oncologic outcome and predictive factors for successful curative salvage surgery after isolated nodal recurrence in head and neck squamous cell carcinoma (HNSCC).
MATERIALS AND METHODS Patient population This study was reviewed and approved by the Hallym University Institutional Review Board (IRB number:
*Corresponding author: Y.-S. Rho, Department of Otorhinolaryngology–Head and Neck Surgery, Ewha Woman’s University, College of Medicine, 911-1 Mok-Dong, Yang Cheon-Gu, Seoul, Korea 134-701. E-mail: [email protected]
KEY WORDS: head and neck cancer, recurrence, surgery, prognosis, mortality
HEAD & NECK—DOI 10.1002/HED
MONTH 2014
1
CHUNG ET AL.
TABLE 1. Pathologic N classifications for the study patients. No. of patients (%) by N classification
Oral cavity Oropharynx Nasopharynx Hypopharynx Larynx Nasal cavity/maxillary sinus Total
N1
N2a
N2b
N2c
N3
Total
4 2 2 1 1 0 10
6 0 0 1 2 2 11
6 2 1 3 6 0 18
3 1 0 0 2 0 6
3 3 0 2 2 0 10
22 8 3 7 13 2 55
The recurrent pathologic N classifications (rpN) are summarized in Table 1. The tumors were classified as rpN1 in 10 patients (18.2%), rpN2a in 11 patients (20%), rpN2b in 18 patients (32.7%), rpN2c in 6 patients (10.9%), and rpN3 in 10 patients (18.2%). Postoperative adjuvant treatment (after salvage surgery) was conducted after multidisciplinary discussion. Surgical treatment was followed by postoperative radiotherapy in 5 patients (9.1%), and chemoradiation in 19 patients (34.5%); among these patients, 16 (66.7%) had previously received radiation. Thirty-one patients (56.4%) were treated with surgery alone. The type of neck dissection depended on the extent of disease and the findings at the time of surgery. Eight patients underwent a bilateral neck dissection (3 therapeutic neck dissections and 5 elective neck dissections). Contralateral elective neck dissections were considered for patients who were node positive at initial treatment, presented with advanced rN classification, and developed recurrence within 1 year of initial treatment.2 Twenty-five patients underwent a radical or modified radical neck dissection. Extended radical neck dissections were performed for 26 patients (extended radical neck dissection was defined as the neck dissection that removes additional lymph node groups or nonlymphatic structures, not ordinarily encompassed by radical neck dissection, and that should be done for the removing of metastatic structure not for the approach to primary site). Carotid artery (n 5 17; carotid subadventitial dissection in 13 patients and carotid resection with/without reconstruction in 4 patients) was the most common structure sacrificed during extended radical neck dissection. Nerves were the next most common (n 5 16; including the hypoglossal, vagus, superior laryngeal, lingual, and phrenic nerve), followed by muscle (n 5 9; including the prevertebral, paraspinal, mylohyoid, stylohyoid, and digastric muscles) and lymph node (n 5 8; retropharyngeal, parapharyngeal, and level VII node; Table 2). Selective neck dissections were performed for 4 patients.
Determination of the optimal cut-point for defining early versus late recurrence The disease-free time was defined as the elapsed time from the first day after the end of curative treatment to the time when recurrence was confirmed. Salvage time was defined as the time after completion of the salvage treatment until recurrence, metastasis, or last follow-up. The 5-year disease-free survival (DFS) and overall 2
HEAD & NECK—DOI 10.1002/HED
MONTH 2014
survival (OS) rates were calculated by using Kaplan– Meier analysis after primary definitive treatment. The optimal cut-point for defining early versus late initial relapse (ie, before salvage) was chosen on the basis of 5year DFS and OS.3
Statistical analysis The OS and DFS rates were determined using the Kaplan–Meier actuarial life-table method with the logrank test for statistical comparison. We used Cox proportional hazard models for the multivariate analysis of survival. A p value < .05 was considered significant.
RESULTS Optimal cutoff value for early versus late recurrence Table 3 summarizes the 5-year DFS and OS rates across different cut-points for initial tumor recurrence before salvage (expressed as disease-free time). The cutoff point of 6 months was chosen because it yielded the most significant difference for the DFS and OS rates.
Surgical outcomes The surgical salvage rate (33 of 55 patients) was 60% for isolated regional recurrence. Recurrent disease developed in 22 patients (40%), with local recurrence in TABLE 2. Type of salvage neck dissection for the study patients. Salvage neck dissection
Selective neck dissection Modified radical neck dissection/radical neck dissection Extended radical neck dissection Lymph node Level VI/level VII Retropharyngeal/parapharyngeal Carotid artery Nerve Hypoglossal Vagus Superior laryngeal Lingual Phrenic Skin Glandular structure Total
No. of patients
4 25 26 8 6 5 17 16 9 6 2 3 3 5 2 55
SALVAGE
TREATMENT FOR ISOLATED REGIONAL RECURRENCES
TABLE 3. Determination of the optimal cut-point for defining early versus late recurrence before salvage surgery. 5-y DFS
5-y OS
Disease-free time, mo
< (%)
(%)
p value
< (%)
(%)
p value
4 5 6 7 8 9 10 11 12 13 14 15 16
72.9 50 52.4 54.5 59.3 58.6 58.8 57.5 57.1 57.1 58.1 58.1 58.1
56.8 64.1 64.7 63.6 60.7 61.5 61.9 66.7 69.2 69.2 66.7 66.7 66.7
.792 .088 .02* .094 .248 .229 .167 .157 .149 .149 .225 .225 .225
45.5 50 52.4 54.5 55.6 51.7 55.9 57.5 57.1 57.1 58.1 58.1 58.1
65.9 66.7 67.6 66.7 67.9 73.1 71.4 73.3 76.9 76.9 75 75 75
.027* .043* .007* .05* .05* .037* .076 .108 .084 .084 .122 .122 .122
Abbreviations: DFS, disease-free survival; OS, overall survival. The figures in bold and * values indicate statistical significance.
TABLE 4. Clinicopathological factors influencing disease-free survival.
Parameter
Previous neck treatment No Yes Initial T classification T1–2 T3–4 Initial N classification N0–1 N2–3 Recurrent N classification N0–1 N2–3 Time to recurrence
