Surgical Salvage for Local and Regional Recurrence in Oral Cancer

J Oral Maxillofac Surg 64:1409-1414, 2006

Surgical Salvage for Local and Regional Recurrence in Oral Cancer Robert A. Ord, DDS, MD, FRCS, MS,* Antonia Kolokythas, DDS,† and Mark A. Reynolds, DDS, PhD‡ Purpose: To evaluate local and regional recurrence and the outcomes for salvage surgery in patients for

oral cancer. Patients and Methods: This study analyzed 354 consecutive patients with oral cancer treated primar-

ily by surgery or surgery combined with adjuvant therapy by 1 surgeon (R.A.O.) between February 1991 and September 2001. Results: Overall recurrence rate was 15.5%; with 5.4% local, 8.5% regional, and 1.4% locoregional. Overall salvage for local recurrence was 52.6% 3-year survival, and statistically significant favorable prognostic factors were salvaged by surgery alone and initial cancer staging of I/II. Overall salvage for regional recurrence was 50%, with recurrence in a previously untreated neck and salvage with radical neck dissection plus radiotherapy giving the best prognosis. No patients with locoregional recurrence were salvaged. Conclusions: Patients who were stage I/II and were treated initially by surgery alone were the best candidates for salvage if they recurred. Salvage was best achieved with surgery or surgery ⫹ adjuvant therapy, and patients recurring within 6 months had a worse survival. Patients with locoregional recurrence or treated with RT ⫾ chemotherapy alone have negligible survival. © 2006 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 64:1409-1414, 2006 The best chance of curing oral cancer is with effective primary treatment. Initial radical therapy has the best success rate, as treatment of recurrent disease has a generally dismal outcome. In view of this, the decision to offer a patient salvage therapy cannot be taken lightly. Factors that must be taken into account are the severity of present symptoms, the prognosis for cure, the functional cost of the proposed salvage therapy and its effects on the patient’s quality of life. In addition, salvage is dependent on the previous therapy the patient has received. Overall, there is a lack of evidence-based data, and most treatment plans

*Professor/Chairman, Department of Oral and Maxillofacial Surgery, University of Maryland, Greenbaum Cancer Center, Baltimore, MD. †Maxillofacial Oncology Fellow, Department of Oral and Maxillofacial Surgery, University of Maryland, Baltimore, MD. ‡Chairman, Department of Periodontology, Baltimore College of Dental Surgery, Baltimore, MD. Address correspondence and reprint requests to Dr Kolokythas: Department of Oral and Maxillofacial Surgery, University of Maryland, 419 West Redwood Street, Suite 410, Baltimore, MD 212011754; e-mail: [email protected] © 2006 American Association of Oral and Maxillofacial Surgeons

0278-2391/06/6409-0015$32.00/0 doi:10.1016/j.joms.2006.05.026

are individualized, involving a multidisciplinary approach. There are many reasons why surgery used for salvage is challenging for both local and regional disease. In local recurrence there may be a delay in diagnosis with deep-seated tumor under reconstruction flaps in scarred and irradiated tissues, making clinical and computed tomography/magnetic resonance (CT/MR) imaging difficult. The advent of the PET scan may improve accuracy in this challenging situation.1,2 Recurrent tumors also have atypical spread with ill-defined margins, and technical problems are further increased by a hypoxic fibrosed bed, after surgery/radiotherapy(RT)/chemotherapy. There may also be a lack of suitable vessels for microvascular reconstruction. In salvage surgery for regional recurrence, difficulties encountered include scarring from previous neck dissection and/or RT, plus delay in diagnosis with increase in number of nodes and extracapsular spread. Reported incidences of local recurrence in oral cancer vary between 12.3% and 29.7%,3-9 and this is related to the status of the resection margin. Positive resection margin rates for oral cancer vary between 8.8% and 39%,7,10-14 however, these figures also reflect differences in definitions of what is a positive margin.

1409

1410

SURGICAL SALVAGE IN ORAL CANCER

Some authors include only invasive tumor at the margin whereas others include carcinoma in situ, dysplasia or tumor within 5 mm of the margin as representing a positive result. Slootweg et al15 found 3.9% recurrence with negative margins and 21.9% with positive margins. Molecular analysis of “negative” histologic margins to conventional light microscopy has shown that they sometimes contain undiagnosed mutated cells, which may explain the phenomena of local recurrence in cases with histologically negative surgical margins.16 This has led to an attempt to develop an intraoperative method to analyze margins at a molecular level.17 However, Slootweg et al’s article15 concluded that the discovery of tumor cells in negative margins by molecular techniques had a negligible clinical impact. Local and regional recurrence remains the commonest cause for failure in oral cancer therapy and salvage surgery results are poor. Goodwin,18 in a meta-analysis of 12 studies on salvage surgery, concluded that there was a 36% 5-year survival rate, however, this study is biased by the inclusion of early-stage laryngeal cancer with a 2-year survival of 84%.18 This current study analyzes a consecutive cohort of oral cancer cases treated at the University of Maryland for local and regional recurrence, and the results of salvage therapy.

Patients and Materials This is a retrospective study of a cohort of consecutive patients with histologically diagnosed epidermoid carcinoma of the oral cavity managed by one surgeon (R.A.O.) between February 1991 and September 2001. Four hundred fifty-two patients were identified initially and inclusion criteria included primary treatment with surgery or surgery plus adjuvant therapy with curative intent, and a minimum follow-up of 36 months or to recurrence/death. One hundred three patients were excluded either because their primary treatment was nonsurgical (chemo/RT); they had persistent disease post-treatment, ie, obvious residual tumor within 4 weeks of completing therapy; they were lost to follow-up, refused all treatment, had

Table 1. RECURRENCE IN 349 ORAL CANCER PATIENTS

Local recurrence Regional recurrence Locoregional recurrence Total

Number of cases

%

19 30 5 54

5.4% 8.5% 1.4% 15.5%

Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

Table 2. MANAGEMENT OF LOCAL RECURRENCE

Surgery (2 ⫹ RT) Chemo/RT RT only Chemo only No therapy Total

Number of Cases

3-yr Survival Postsalvage

%

12 1 2 2 2 19

9 1 0 0 0 10

75% 100% 0% 0% 0% 52.6%

Abbreviation: RT, radiotherapy. Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

been previously treated for this cancer, or were inadequately documented. The resultant study group thus consisted of 349 previously untreated oral cancer patients treated with “gold standard” primary surgery or surgery/RT for oral cavity cancer. Data recorded and analyzed included stage of primary tumor, site of recurrence, time to recurrence, type of salvage therapy, and survival postsalvage. Data were statistically analyzed using Fisher’s exact test, where appropriate, to compare survival outcome.

Results In 349 oral cancer patients treated primarily with surgery, there were 54 patients who developed recurrence during follow-up (15.5%) (Table 1). Local recurrence occurred in 19 patients (5.4%), regional recurrence in 30 patients (8.5%), and locoregional recurrence in 5 patients 1.4%. The local recurrences were diagnosed at a mean of 21 months, with 7 of 19 diagnosed at ⬍1 year. Seven of 19 were considered unresectable, and only 1 of these patients was salvaged with combination chemo/RT. Twelve patients were able to undergo salvage surgery, 2 with adjuvant RT, and 9 of these patients were salvaged (survived ⬍3 years) (Table 2). Overall 3-year survival was 52.6% postsalvage therapy, with only 1 patient surviving who did not receive surgery (P ⱕ .05). Factors prognostic for better salvage were primary treatment with surgery only and initial diagnosis of T1/2 (P ⱕ .05) (Tables 3, 4). Survival experience of patients receiving surgery alone was 80%, compared with 22% for those undergoing surgery with adjuvant RT or combination chemo/RT (P ⱕ .05). The mean time to regional recurrence was 9 months, and 23 of 30 were diagnosed at ⬍1 year. Nineteen of these recurrences occurred in a previously untreated neck, 13 were ipsilateral, 4 contralateral, and 2 bilateral. Eleven regional recurrences occurred in previously treated necks, 3 ipsilateral within the levels encompassed by the previous neck dissec-

1411

ORD ET AL

Table 5. REGIONAL FAILURE AND SALVAGE

Table 3. SALVAGE FOR LOCAL RECURRENCE

Primary Therapy

Number of Cases

Number Salvaged

(%)

Surgery Surgery ⫹ RT Surgery ⫹ Chemo/RT Total

10 8 1 19

8 1 1 10

80% 12.5% 100% 52.6%

Abbreviation: RT, radiotherapy. Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

Patient Number

Number Salvaged

%

19 3 3 3 1 1 30

13 1 1 0 1 0 15

68% 33% 33% 0% 0% 0% 50%

Previously untreated neck Recurred within ND Recurred within RT Recurred outside ND Recurred outside ND/RT Recurred bilateral within ND Total

Abbreviations: ND, neck dissection; RT, radiotherapy.

tion, 3 ipsilateral outside the levels previously dissected, 3 ipsilateral in an irradiated neck, 1 ipsilateral outside the levels previously dissected and irradiated, and 1 bilateral within the levels previously dissected. A total of 15 of 30 patients (50%) were salvaged (Table 5). The average survival time for patients with regional recurrence who failed salvage was 8.8 months. The best results for salvage were given by (modified) radical neck dissection plus RT in a previously undissected neck (P ⱕ .08). Salvage rates fell in patients who had any form of previous therapy and no patient was salvaged without surgery as part of their treatment. The treatments and results for this group are shown in Table 6. There were 5 patients diagnosed with locoregional recurrence and all succumbed to their disease with zero salvage. The average time to recurrence was 8.8 months and the average time of survival was 5 months. Primary therapy for these 5 patients was surgery plus RT for 4 patients and surgery alone for 1 patient. Only 1 of these patients was believed to be resectable for recurrent disease, and at the time of surgery his tumor was found to be more extensive than clinically indicated and clean margins could not be obtained. The treatments and results for this group are shown in Table 7.

Discussion The literature indicates that salvage surgery is the only therapy that offers any meaningful survival for recurrent disease. Wong et al19 reviewed 377 recurrent head and neck tumors, 121 (32%) of which were

Table 4. SALVAGE FOR LOCAL RECURRENCE

Initial Stage

Patient Number

Number Salvaged

%

I/II III/IV

8 11

7 3

87.5% 27.2%

Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

primary oral cavity cancers. They found that patients treated with salvage surgery had an overall 5-year survival of 26% with a mean survival of 29 months. In contrast, the 5-year survival for both RT and chemotherapy was 0% with a mean of 7 months (little different for supportive therapy only), with a 0% 5-year survival and a mean of 5 months. Patients with local recurrence had a 35% 5-year survival with a mean of 33 months, for regional recurrence there was a 26% 5-year and 28-month mean. However, for patients with locoregional recurrence there was a 0% 5-year survival and a mean of only 8 months. This study showed the value of surgery in salvage but also that patients with locoregional disease may not be salvageable. The factors determining success rates for salvage in local recurrence have been reported by a number of authors. Schwartz et al20 reviewed 350 patients with oral cancer, 99 (28%) recurred and only 38 of these could be treated with curative intent; 92% of recurrences occurred within 36 months. Local recurrence was the most common at 58%, followed by locoregional at 28% and regional 16.5%. Patients that recurred locally within 6 months had a 0% salvage rate with a mean survival time of 20 months, whereas patients who recurred ⬎6 months had a mean survival of 54 months, with 8 patients salvaged. Their

Table 6. SALVAGE THERAPY FOR REGIONAL RECURRENCE

Radical ND Radical ND ⫹ RT Radical ND ⫹ RT/Chemo Chemotherapy No treatment Total

Patient Number

Number Salvaged

%

7 17 1 1 4 30

2 12 1 0 0 15

28.5% 70.6% 100% 0% 0% 50%

Abbreviations: ND, neck dissection; RT, radiotherapy. Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

1412

SURGICAL SALVAGE IN ORAL CANCER

Table 7. TREATMENT AND SALVAGE FOR LOCOREGIONAL RECURRENCE

Surgery ⫹ RT Chemotherapy No therapy Total

Patient Number

Salvage

%

1 3 1 5

0 0 0 0

0% 0% 0% 0%

Ord et al. Surgical Salvage in Oral Cancer. J Oral Maxillofac Surg 2006.

overall salvage cure rate was 21%. Patients who had Stage I/II primary tumors had significantly improved salvage time and total survival time compared to patients with primary tumors Stage III/IV. However, these authors found that the stage of the recurrent tumor made no difference to the salvage rate and recommended against restaging recurrences to assess prognosis. Again in this series there was a significantly improved salvage time and survival time with patients who had salvage surgery compared with patients who had RT plus or minus chemotherapy. But salvage cure rate was not influenced by the stage of the primary or recurrent tumor, or the type of salvage treatment statistically. Eckhardt et al7 studied 1,000 patients, 88% with oral cancer, the vast majority being treated with primary surgery or surgery plus RT; 198 (19.8%) recurred and 79% of these were within 2 years. The 5-year survival for recurrent disease was 14.4%. Patients with primary tumors T1/2 had a lower incidence of recurrence (26.4%) than those with primary tumors T3/4 (44.6%). Salvage surgery was performed in 65 cases, with a 31% 5-year survival, salvage surgery plus adjuvant therapy in 52 cases with a 15.4% 5-year survival, 42 patients had adjuvant therapy alone with 0% 5-year survival, and the 39 patients with supportive therapy alone had a 0% 5-year survival. Although the differences for survival with surgery alone compared with other treatment and surgery plus adjuvant therapy compared with adjuvant alone were highly statistically significant, there are obvious biases inherent in this study. Patients who required adjuvant therapy in addition to surgery would be expected to have a more advanced stage or poor prognostic factors such as positive margins or perineural spread compared with the group having surgery alone. Similarly, the group with adjuvant therapy only probably contains the most advanced unresectable disease or patients too medically compromised to undergo major surgery. These biases are found in all articles published on salvage, so the striking advantage displayed by surgery must be viewed in the light that the most prognostically favorable patients are being treated surgically. Lin et al21 examined 191 patients with oral cancer, comparing 135 with no previous therapy to 53 with

recurrent disease to see whether recurrent disease indeed had such a poor prognosis. They staged both groups as early or late (essentially stage I/II and stage III/IV). As would be expected, overall absolute and disease-free survival for the untreated group at 84% and 48%, respectively, was superior to that of the recurrent group, 48% and 24%, respectively. However, absolute survival in early and late untreated was 92% and 64%, respectively, compared with 60% and 38%, respectively, in the recurrent group. These results are mirrored in the 5-year disease-free survival in early and late untreated, 62% and 26%, respectively, compared with the recurrent group, 32% and 16% respectively. Interestingly, this study showed that early recurrent disease had a better prognosis than late and that perhaps restaging recurrent disease is useful. Also, the data would indicate that early recurrent disease had a very similar outcome to untreated patients with late stage disease. Lin et al21 also compared the complications between the 2 groups, finding, not surprisingly, that the recurrent group of patients had a significantly higher rate of overall complications, 60.7% versus 30.4%. However, there was no significant difference in major complications or in patients who had or had not received RT. Gleich et al22 attempted to look at a group of head and neck cancer patients who should have the worst prognosis for recurrence, reporting 940 patients who were initially T3/4. Forty-eight patients recurred and had salvage therapy, 24 with local and 4 with locoregional recurrence. Twenty-five had surgery (2 ⫹ RT) and 3 had RT alone. Only 2 patients were alive (8.5%) and 3 others survived more than 2 years. Only 1 patient who had “gold standard” initial therapy of surgery plus RT was salvaged. He also confirmed Schwartz et al’s work20 that there was no relationship between the stage of recurrent tumor and survival. In comparing our own data, we have a very low 5.4% incidence of local recurrence. This is probably a reflection of our exclusion criteria that many stage III/IV patients with advanced or unresectable disease were treated with chemo/RT, as were patients with significant comorbid disease who were medically considered inoperable. In addition, we did not include persistent disease. Our cohort was therefore a highly selected group treated with primary surgery for cure. In this situation it would be reasonable to expect a lower incidence of recurrence. Our findings support the use of surgery as the best means of salvaging local recurrence. If the patient is resectable, salvage surgery does offer a survival benefit. We were also able to confirm the findings that patients with initial diagnosis stage I/II had a better prognosis for salvage than those with stage III/IV. In addition, as for other authors, our results indicate that patients treated with surgery alone had a better survival with salvage than

ORD ET AL

those treated with multimodality therapy. However, it is difficult to tell whether this is an independent variable, as it is patients with stage III/IV that will receive multimodality therapy and this may be a “stage effect.” Our salvage rate of 52.6% is high and this is due to the fact that most of our patients were suitable for resection and that our recurrences were mostly marginal rather than deep. This may be because many of our primary tumors were in the tongue, and with most of our patients having primary microvascular reconstruction, our deep margins are generous (1.5-2 cm). We would like to avoid deep-seated recurrence, which is difficult to salvage, and we are able to make bigger defects with good functional outcomes with free flap reconstruction. In reporting their experience with regional recurrence, Wong et al19 found that patients with ipsilateral recurrence in a previously unoperated neck had a 56% surgical salvage with a 5-year survival of 32%. Whereas in ipsilateral recurrence in a patient with a previous neck dissection these figures fell to 32.5% surgical salvage and a 5-year survival of 18%. There was 0% successful salvage with RT or chemotherapy. Kowalski,23 in a series of 513 patients with oral cancer, had 82 (16%) with regional recurrence, 94% of which recurred within 2 years. Salvage treatment could be attempted on only 51 patients (62%). In his 82 regional recurrences, 44 were ipsilateral, 31 contralateral, and 7 bilateral. Only 5 of 46 (10.8%) patients who had recurrence in an unoperated neck survived. Factors that reached significance in deciding prognosis after salvage neck dissection were the type of previous neck dissection, whether the patient had previous RT, and also those patients who recurred in ⬍6 months had a statistically better survival than those who recurred ⬎24 months. This last result is counter to the rest of the literature, which states that the earlier the recurrence, the worse the prognosis. Sun et al24 reported 233 cases of tongue cancer, 45 of which recurred regionally (19.3%). The salvage rate was 26%. In univariate analysis, neck relapse had a significantly worse prognosis than local recurrence. Patients who relapsed in ⬍6 months had a worse prognosis and there was a trend toward better results form surgery as salvage. Mabanta et al25 reviewed 356 head and neck patients with clinically positive nodes treated primarily with RT. Fifty-one (14%) recurred in the neck with a mean time of 11.2 months to recurrence. Thirty-three (65%) of these patients had no salvage therapy, and 18 underwent salvage, 14 with surgery (2 ⫹ RT, 1 ⫹ chemo), 4 with chemotherapy alone. All patients developed cancer at local, regional, distant or a combination of these sites. The 5-year regional control was 9% and the salvage group 3-year absolute and cause-specific survival was 44%, compared with 0% for the nonsalvage group. These au-

1413 thors concluded that the likelihood of successful salvage for regional recurrence post-RT is remote. Gleich et al,22 in their patients initially T3/4, had 18 evaluable patients with regional recurrence. Four patients were treated with RT alone and persisted/recurred within 6 months. Fourteen patients had neck dissection with a mean survival time of 31.1 months, and 6 of these patients had prolonged survival. No patient who had previous neck dissection plus RT was salvaged. In our own series we had 30 regional recurrences (8.5%), and although this is a low figure, it is more than our local recurrence rate, perhaps reflecting that we were less aggressive with elective neck dissection in the N0 neck at the start of this series 12 to 15 years ago. Our results again show best results in previously untreated necks and our high salvage rate is due to the fact that 19 of 30 (63.3%) of our recurrences were in previously untreated necks. Salvage in patients who have had surgery or RT in the neck is significantly worse. No patient was salvaged without surgery and modified radical neck dissection plus RT gave the best results. There are few articles that analyze locoregional recurrence separately, although Wong et al19 Sun et al,24 and our own study have shown a salvage rate of 0. Perhaps the inclusion of those patients with such a terrible prognosis in the local or regional recurrence groups in some articles lowers the overall salvage rates, giving a falsely dismal picture. It is hard to know whether patients who have locoregional recurrence have very aggressive cancers or whether these patients have more residual disease post initial therapy. These patients may be incurable and should always be analyzed separately. In our study, albeit with a selected cohort, overall salvage rates for local recurrence of 52.6% and 50% for regional recurrence support the judicious use of aggressive surgical salvage. Conclusions that can be drawn from recent studies and our own experience are: Surgery plus or minus adjuvant therapy is the best modality for salvage therapy. Currently RT and chemotherapy appear to have little advantage over supportive care, although these modalities have been used to treat those patients least likely to be cured. Patients who had early-stage disease initially are most likely to be salvaged. T3/4 cancers have a higher rate of local recurrence, more morbidity, and a lesser chance of salvage. Most studies do not support the value of restaging recurrent disease in head and neck cancer, finding no difference in salvageability between early and late recurrence. Patients treated initially with surgery alone are the best candidates for salvage, as previous RT and/or

1414 chemotherapy make salvage less likely. The more modalities that were used in primary therapy, the worse the prognosis for recurrent disease. However, this may also be due to the fact that the more advanced the initial disease, the more likely it is to require multimodality therapy. It is likely that with the recently published Radiation Therapy Oncology Group (RTOG)26 and European Organization for Research and Treatment of Cancer (EORTC)27 studies, more patients will receive chemo/RT as opposed to RT alone if they have poor prognostic factors. Thus, we may see increased initial local and regional control with decreased ability to salvage recurrence. Almost all local and regional recurrence occurs within 3 years, and early recurrence ⬍6 months carries a worse prognosis. Locoregional recurrence is probably incurable.

References 1. Ware RE, Matthews JP, Hicks RJ, et al: Usefulness of fluorine-18 fluorodeoxyglucose positron emission tomography in patients with residual structural abnormality after definitive treatment for squamous cell carcinoma of the head and neck. Head Neck 26:1008, 2004 2. Terhaard CH, Bongers V, van Rijk PP, et al: F-18-fluoro-deoxyglucose positron emission tomography scanning in detection of local recurrence after radiotherapy for laryngeal/pharyngeal cancer. Head Neck 23:933, 2001 3. Leemans CR, Tiwari R, Nauta JJP, et al: Recurrence at the primary site in head and neck cancer and the significance of lymph node metastasis as a prognostic factor. Cancer 73:187, 1994 4. Cooney TR, Poulsen MG: Is routine follow-up useful after combined-modality therapy for advanced head and neck cancer? Arch Otolaryngol Head Neck Surg 125:379, 1999 5. Kowalski LP, Magrin J, Waksman G, et al: Supraomohoid neck dissection in the treatment of head and neck tumors: Survival results in 212 cases. Arch Otolaryngol Head Neck Surg 119: 958, 1993 6. Magnano M, Bussi M, De Stefani A, et al: Prognostic factors for head and neck tumor recurrence. Acta Otolaryngol 115:833, 1995 7. Eckardt A, Barth EL, Kokemueller H, et al: Recurrent carcinoma of the head and neck: Treatment strategies and survival analysis in a 20-year period. Oral Oncol 40:427, 2004 8. Hemprich A, Muller RP: Long-term results in treating squamous cell carcinoma of the lip, oral cavity, and oropharynx. Int J Oral Maxillofac Surg 18:39, 1989 9. Rodrigo JP, Suarez C: Prognostic significance of postoperative wound infection on head and neck cancer. Otolaryngol Head Neck Surg 118:272, 1998

SURGICAL SALVAGE IN ORAL CANCER 10. Zieske LA, Johnson JT, Myers EN, et al: Squamous cell carcinoma with positive margins. Surgery and postoperative irradiation. Arch Otolaryngol Head Neck Surg 112:863, 1986 11. Cook JA, Jones AS, Phillips DE, et al: Implications of tumour in resection margins following surgical treatment of squamous cell carcinoma of the head and neck. Clin Otolaryngol Allied Sci 18:37, 1993 12. Jones AS, Hanafi BZ, Nadapalau V, et al: Do positive resection margin after ablative surgery for head and neck cancer adversely affect prognosis? A study of 352 patients with recurrent cancer following radiotherapy treated by salvage surgery. Br J Cancer 74:128, 1996 13. Ord RA, Aisner S: Accuracy of frozen sections in accessing margins in oral cancer resection. J Oral Maxillofac Surg 55:663, 1997 14. Woolgar LA, Scott J, Vaughan ED, et al: Survival, metastasis and recurrence of oral cancer in relation to pathological features. Ann R Coll Surg Engl 77:325, 1995 15. Slootweg PJ, Hordijk GJ, Schade Y, et al: Treatment failure and margin status in head and neck cancer. A critical view of the potential value of molecular pathology. Oral Oncol 38:500, 2002 16. Brennan JA, Mao L, Hruban RH, et al: Molecular assessment of histopathological staging in squamous-cell carcinoma of the head and neck. N Engl J Med 332:429, 1995 17. Goldenberg D, Harden S, Masayesva BG, et al: Intra-operative molecular margin analysis in head and neck cancer. Arch Otolaryngol Head Neck Surg 130:39, 2004 18. Goodwin WJ Jr: Salvage surgery for patients with recurrent squamous cell carcinoma of the upper digestive tract: when do the ends justify the means? Laryngoscope 110:1, 2000 19. Wong LY, Wei WI, Lam LK, et al: Salvage of recurrent head and neck squamous cell carcinoma after primary curative surgery. Head Neck 25:953, 2003 20. Schwartz G, Mehta R, Wenig B, et al: Salvage treatment for recurrent squamous cell carcinoma of the oral cavity. Head Neck 22:34, 2000 21. Lin Y-C, Hsiao J-R, Tsai S-T: Salvage surgery as primary treatment for recurrent oral squamous cell carcinoma. Oral Oncol 40:183, 2004 22. Gleich LL, Ryzenman J, Gluckman J, et al: Recurrent advanced (T3 or T4) head and neck squamous cell carcinoma. Is salvage possible? Arch Otolaryngol Head Neck Surg 130:35, 2004 23. Kowalski LP: Results of salvage treatment of the neck in patients with oral cancer. Arch Otolaryngol Head Neck Surg 128:58, 2002 24. Sun LM, Leung S, Su C, et al: The relapse patterns and outcome of postoperative recurrent tongue cancer. J Oral Maxillofac Surg 55:827, 1997 25. Mabanta S, Mendenhall W, Stringer S, et al: Salvage treatment for neck recurrence after irradiation alone for head and neck squamous cell carcinoma with clinically positive neck nodes. Head Neck 21:591, 1999 26. Cooper JS, Pajak TF, Forastiere AA, et al: Post-operative concurrent radiotherapy and chemotherapy for high risk squamous cell carcinoma of the head and neck. N Engl J Med 350:1937, 2004 27. Bernier J, Domenge C, Ozsahin M, et al: Post-operative irradiation with or without concomitant chemotherapy for locally advanced head and neck cancer. N Engl J Med 350:1945, 2004