Can neck irradiation be an alternative to neck dissection in early stage carcinoma oral tongue operated for primary alone? Experience from a single institute

Purpose: To study pattern of failure, locoregional control rates (LCR) and disease free survival (DFS) in post-operative patients of carcinoma oral tongue, and to study the impact of nodal dissection on DFS in stage I and II patients. Methods: 102 patients of carcinoma oral tongue treated between January 2009 and December 2013 were analyzed. All patients were operated for primary disease, but neck dissection was done in 78 (76.5%) patients only. However, radiation to primary site along with neck region was received by all patients. Pattern of failure, LCR and DFS were estimated. Results: At median follow up of 12 months, 10.8% patients failed locally, 10.8% in nodal region, 2.9% both at local and nodal site, and 5.9% patients failed distally. 2 year LCR and DFS was 71.2%, 90.9%, 79.5%, 0% and 55.2%, 64.4%, 57.8%, 0% in stage I, II, III, IV respectively. 2 year DFS in stage I patients, who underwent nodal dissection and post-operative radiation (14 patients) was 64.3% and in whom only neck irradiation was done (15 patients), it was 45.8%, however difference was not significant ( p = 0.5). But in stage II patients, 33 patients who underwent nodal dissection and post-operative radiation, 2 year DFS was 85.4% and it was 21.4% in 7 patients who underwent neck radiation only, and difference showed trend towards significance ( p = 0.05). 2 or more positive lymph nodes post dissection was the only poor prognostic factor that correlated with DFS ( p = 0.02) Conclusion: While in stage I, neck irradiation alone can be a possible alternative to neck dissection and post-operative radiation; for stage II, neck dissection is mandatory.


Introduction
Tongue cancer accounts for 25 to 40% of oral squamous cell carcinomas. 1 Being a highly muscularized structure and having rich lymphatic network, tongue cancer is predisposed for early lymph node metastasis. As such, lymph node metastasis, both occult and manifest, are observed more commonly in oral tongue cancer than in any other cancer of the oral cavity. 2 Over the years, the management of early stage tongue cancer (clinical tumor classification [cT] cT1 -T2 N0 M0), has seen a major change, both for the primary local disease, as well as for the neck nodes. Interstitial brachytherapy has been used widely for early stage disease, in place of wild local excision or partial glossectomy, in an attempt to preserve the organ. However elective neck dissection in early stage disease, has been a source of debate in recent years. 3, 4

Introduction
Tongue cancer accounts for 25 to 40% of oral squamous cell carcinomas. 1 Being a highly muscularized structure and having rich lymphatic network, tongue cancer is predisposed for early lymph node metastasis. As such, lymph node metastasis, both occult and manifest, are observed more commonly in oral tongue cancer than in any other cancer of the oral cavity. 2 Over the years, the management of early stage tongue cancer (clinical tumor classification [cT] cT1 -T2 N0 M0), has seen a major change, both for the primary local disease, as well as for the neck nodes. Interstitial brachytherapy has been used widely for early stage disease, in place of wild local excision or partial glossectomy, in an attempt to preserve the organ. However elective neck dissection in early stage disease, has been a source of debate in recent years. 3, 4 Spiro and Strong analyzed 314 patients (1957 -1963) of tongue cancer, in whom neck dissection was not done, and demonstrated an overall 5-year survival rate of 42% International Journal of Cancer Therapy and Oncology www.ijcto.org

Introduction
Tongue cancer accounts for 25 to 40% of oral squamous cell carcinomas. 1 Being a highly muscularized structure and having rich lymphatic network, tongue cancer is predisposed for early lymph node metastasis. As such, lymph node metastasis, both occult and manifest, are observed more commonly in oral tongue cancer than in any other cancer of the oral cavity. 2 Over the years, the management of early stage tongue cancer (clinical tumor classification [cT] cT1 -T2 N0 M0), has seen a major change, both for the primary local disease, as well as for the neck nodes. Interstitial brachytherapy has been used widely for early stage disease, in place of wild local excision or partial glossectomy, in an attempt to preserve the organ. However elective neck dissection in early stage disease, has been a source of debate in recent years. 3, 4 Spiro and Strong analyzed 314 patients (1957 -1963) of tongue cancer, in whom neck dissection was not done, and demonstrated an overall 5-year survival rate of 42% only. 5 Franceschi et al. in a study conducted in 297 patients (1978)(1979)(1980)(1981)(1982)(1983)(1984)(1985)(1986)(1987) of oral tongue cancer demonstrated an improved overall 5-year survival rate of 65%, which was related to a more aggressive treatment of the neck even in early tumour stages and to adjuvant radiotherapy in advanced tumour stages. 6 However, there is limited literature on the efficacy of neck irradiation in controlling the occult lymph node metastasis from early stage carcinoma tongue, when neck dissection has not been done. Most of the data can be extrapolated from the survival benefits achieved by chemoradiation in patients with carcinoma oropharynx, where neck irradiation has shown results similar to nodal dissection in controlling the neck disease. 7 The purpose of the present study was to describe our experience with surgical based therapy of tongue cancer during the last five years. Also, whether in early stage patients (stage I and II), neck dissection can be replaced by neck irradiation or not, has been discussed. Furthermore, prognostic factors for survival were analyzed in order to obtain valid criteria for therapeutic decision-making in clinical routine.

Patients
Between January 2009 and December 2013, one hundred and two post-operative patients of stage I -IV A carcinoma tongue, registered and further treated at our radiation oncology department were analyzed. Patients had Karnofsky performance status > 70 and adequate haematologic (haemoglobin > 10 gm/dl, absolute neutrophil count > 1500/dl, platelets > 100,000/dl), hepatic and renal function (calculated creatinine clearance > 60 mL/min). Exclusion criteria included inoperable cases due to disease extension or medical comorbidities, stage IV -B disease, previous treatment with RT or chemotherapy, any prior or synchronous malignancy, hypersensitivity to platinum agents and serious medical disease or pregnant state.

Surgery type
All 102 patients underwent surgery for local primary disease, wide local excision (WLE) (56 patients), hemiglossectomy (33), total glossectomy (2), WLE plus segmental mandibulectomy (10), total glossectomy plus segmental mandibulectomy (1); but only 78 (76.5%) patients out of them had underwent neck dissection also, for nodal control. The neck dissection done was ipsilateral (I/L) radical neck dissection (RND) in 15 patients, I/L modified neck dissection (MND) in 20, I/L supraomohyoid neck dissection (SOND) in 32, I/L RND plus contralateral (C/L) MND / SOND in 2 , and I/L MND plus C/L MND/SOND in 9 patients. It is worth mentioning that in the Otolaryngology department of our institute, all patients of early stage (I and II) oral tongue cancer are managed by elective neck dissection along with the treatment of the primary disease. But 26 patients (25.5%) already had undergone surgery for primary disease outside the institute, and the neck was not addressed in 24 patients out of them (all these 24 patients were early stage). Such patients when referred to our institute, are not sent for second surgery in the form of neck dissection, but are managed by elective neck irradiation at our department of radiation oncology.
Therefore, all patients irrespective of neck dissection done or not, were planned with local radiation to primary as well as neck region post operatively. Indication for giving post-operative radiation in these patients was inadequate dissection, pathological pT2, 3, 4, high risk features like poor differentiation, margin positivity, depth of invasion > 4mm, extracapsular extension and node positivity. All the clinically node negative early stage (I, II) patients (T1, 2 N0 M0) who underwent surgery for primary disease alone, were given elective neck irradiation, considering non neck addressal to be inadequate surgery in these patients.

Radiation technique and dose
Patients were simulated on Simulator CT (Phebus Mecaserto, France) after immobilisation with a thermoplastic mould and treated with either Co -60 crays or 6 MV photons. Patients were treated by parallel-opposed lateral portals without any tissue compensators. Neck nodes were treated electively in all patients who received external radiation. Regarding dose, 40 Gy was delivered in 20 fractions in 4 weeks to the primary and draining lymph nodes (phase I), followed by 10 Gy in 5 fractions in 1 week after sparing the spinal cord (phase II). Additional 10 Gy in 5 fractions in 1 week was given in the presence of high risk features. Therefore, the dose up to 50 Gy in 25 fractions was given to nodal region only in those patients, who were early stage I and II clinically and also had pathologically uninvolved necks. Rest all patients were planned up to 60 Gy in 30 fractions.
Patients with 2 or more lymph nose positivity and those with extracapsular extension were planned with chemoradiation (CRT). In these patients, concurrent single agent cisplatin, 100 mg/m 2 intravenously was administered on days 1, 22 and 43 of the radiation schedule after proper hydration. Radiation was administered within 2h after the cisplatin administration. A complete haemogram and renal function tests were done before every cycle of cisplatin. Chemotherapy was withheld in cases of any grade 2 or more haematologic or renal toxicity, till the normal values were recovered after specific management. two months for the first year and then quarterly. Chest X-rays were obtained at 6 months intervals. Recurrence at local or nodal site was considered as local or regional failure from day zero. Fine needle aspiration cytology or a biopsy was carried out to document a recurrence in clinically suspicious cases.

Statistical analysis
In this retrospective study, frequency tables with counts and percentages were used to describe pre-treatment and treatment characteristics of the patients. Actuarial disease free survival (DFS) and overall survival (OS) rates were calculated by the Kaplan -Meier method and stratified by stage groups. The survivals were compared between early stage patients on the basis of nodal dissection, using log -rank test. Exploratory subgroup analysis was carried out on various prognostic variables. The relationship between the clinic pathologic variables and survival was assessed in univariate analysis using the log rank test. For multivariate analysis, the Cox proportional hazard model was used. A p-value of < 0.05 was taken as significant. Data were analyzed using the statistical software SPSS for windows (version 19.0). Table 1 shows the profile and treatment details of the treated patients. Out of 102 patients with median age of 48 years (range: 24-83 years), 29 (28.4%) patients were stage I, 40 (39.2%) were stage II, 25 (24.5%) were stage III, and 8 (7.8%) were stage IVa. 74.5% of the total patients were operated at our institute, rest were operated outside. All 102 patients underwent surgery for local primary disease, but only 78 (76.5%) patients out of them had underwent neck dissection also, for nodal control. All patients however had received postoperative radiation to the primary as well as neck region.

Early stage, clinically node negative, but pathologically node positive
Out of 29 stage I patients, only 1 (3.4%) was pathologically node positive, but out of 40 stage II patients, 16 out of 33 patients who underwent neck dissection (40% of stage II) were pathologically node positive. This clearly indicates the definite need for neck addressal even in early stage patients. Table 2 shows the postoperative histopathological features among the patients.

Pattern of failure
At 12 month median follow up period, 31 (30.4%) patients failed. Table 3 shows the pattern of failure among these patients and the stage to which they belong.
However, p N status was significantly correlated to nodal disease free survival in the univariate analysis (p = 0.03). However, in multivariate analysis (Cox proportional hazard model), the only variable which had significant correlation to the disease free survival was the number of positive lymph nodes > 2 post dissection (p = 0.02).

Attempt for salvage therapy
All recurrences were verified histologically, unless obvious by clinical examination. In patients with residual tumour, disease recurrence, or progression of disease, salvage surgery or palliative treatment was offered, depending on the status of the individual patient, their symptoms and previous treatment.
Among the 11 local recurrences, 6 patients were planned for total glossectomy, but only 4 underwent the same (3 patients out of these also received radiation post surgery) and rest were lost to follow up. 5 patients were planned for palliative chemotherapy, but only 1 could receive the same, rest were lost to follow up.
Among the 11 patients who had isolated nodal failures, 2 patients underwent radical neck dissection, 4 patients who either already had neck dissection or had inoperable lymph node relapse, received radiation (with or without chemotherapy), 1 patient was planned with palliative chemotherapy and rest were lost to follow up.
Out of 6 patients who had both local and lymph node relapse, only 2 received palliative chemotherapy, and rest were suitable for best supportive care only. Out of 3 patients with distant relapse, only one received palliative chemotherapy, 2 others were given best supportive care.

Discussion
Tongue is the most common subsite for squamous cell carcinoma of oral cavity, accounting for 25 -40% of total oral cancers. 1 Among the total cases of carcinoma oral tongue, 30% are early stage and nearly 70% are advanced stage. 2,3 Despite the development of multimodal treatment options, the prognosis remains relatively poor.
In a retrospective study, conducted by Kokemueller in 341 patients with squamous cell carcinoma of the tongue, between 1980 and 2009, it was found that local and regional failures occurred in 23.9% and 20.4% patients respectively, leading to a total failure rate of 37.2% after an average duration of 1.6 years. 8  In our study, pN status (p = 0.07) and neck dissection (p = 0.08) showed trend towards significance in univariate analysis (log rank), and were identified as prognostic factors for disease free survival. p N status was also significantly correlated to nodal disease free survival in the univariate analysis (p = 0.03). In multivariate analysis, the only variable which had significant correlation to the disease free survival was the number of positive lymph nodes > 2 post dissection (p = 0.02) The overall locoregional recurrence rates as described by the studies in literature range between 16 and 42%. 11 -16 But in our study, stage I patients had higher local recurrences after wide local excisions, as compared to the other stages, which is unusual. It was found out that all these patients had undergone surgery for the primary disease from outside the institute. Therefore the comment could not be made on the adequacy of dissection. This is also the reason why overall local control rates and disease free survivals are lesser in stage I compared to stage II. Similarly, the overall survival rate of our patients with tongue cancer is lesser than the survival rates described by other authors, which are quoted between 40 and 65%. 5, 6, 17 Early stage carcinoma tongue patients (T1 T2 N0M0) generally have good survival. Ganly et al. showed that 5-year disease -specific and overall survival rate was 86% and 79% respectively. 18 The management for such patients includes treatment for the primary site and that of the cervical lymph nodes. For the treatment of primary local site, surgery and brachytherapy are the widely used options. However, the management of patients with clinically negative nodes (N0) with early tongue cancer is controversial. There has been a debate whether such patients should be kept on follow up or should undergo elective neck dissection or only local radiation to neck without neck dissection would suffice. Those who favor wait and watch policy argue that 80% of patients with N0 neck would be over treated, and subjected to additional morbidity and costs. Weiss et al. suggested that elective neck dissection is necessary if the incidence of occult metastasis is greater than 20%. 19 It is generally accepted that cancer of the oral tongue often shows lymph node involvement even in early stages .The proportion of occult metastases is quoted between 24 and 42%. 20, 21, 22 Even in our study, 40% of the stage II patients who were clinically node negative, were found to be pathologically node positive after neck dissection.
Two randomized controlled trials (RCTs) performed to compare elective neck dissection with observation alone, did not find any survival difference between the two arms and tumor depth of > 4 mm was associated with higher rates of involved nodes and suggested that these set of patients may benefit from elective neck dissection. 23,24 However, both these RCTs had small numbers and consisted of methodology flaws, so their results could not be applied clinically.
In our part of the world, where patient compliance for follow up is poor, "wait and watch" policy for early stage carcinoma tongue patients (operated for primary alone), can be disastrous. Therefore, all these patients are taken up for post-operative radiation to primary and neck at our institute.
Till date, there is no study in literature which has prospectively compared elective neck dissection to neck irradiation alone in early stage patients of carcinoma tongue. However there are some retrospective studies which have analyzed the role of post-operative radiotherapy pT1 -T2 N0 deep tongue cancers. Gokavarapu et al. analyzed 103 patients primary pT1 -T2 N0 oral tongue cancer of depth of invasion 4 mm or greater treated surgically from January 2010 to December 2012. 25 62 patients received post-operative radiotherapy (PORT) and 41 patients did not receive the same; median period of follow -up was 41.3 months. Logistic and Cox regression models showed no significant difference in locoregional recurrences (P = .078) and survival (P = 0.339) between patients who received PORT and those who did not receive PORT.
In the present study also, no significant impact of depth of invasion was found on the loco -regional recurrences and survival. However, by determining stage wise locoregional and disease free survival rates, the present study was able to find out whether neck irradiation alone could replace the elective neck dissection and postoperative radiation in early stage I and II carcinoma tongue patients.
Our results show that out of 29 clinically node negative stage I patients, only 1 was pathologically positive. The 2 year disease -free survival in stage I post-operative patients of carcinoma tongue, who underwent nodal dissection and post-operative radiation was 64.3% and in those in whom nodal dissection was not done, but radiation was given, it was 45.8%, and the difference was not statistically significant (p = 0.5). This indicates that neck addressal is mandatory, however neck irradiation can be considered as an alternative to neck dissection for stage I patients of carcinoma tongue. But for stage II patients, it was found that 2 year DFS was higher in patients who had undergone neck dissection and post-operative radiotherapy, compared to neck irradiation alone, and p value of 0.05 shows trend towards significance.
However, the basic limitation of this study is that this early stage carcinoma tongue group of patients does not represent the entire early stage carcinoma tongue patients being treated at our institute. There are some patients with stage T1 N0M0 who underwent surgery for local primary with or without neck dissection under otolaryngology department, and the histopathology being favorable, were kept on follow up, and hence were not accessible for the present study. Therefore, without analyzing that subgroup of early stage carcinoma tongue patients, it is difficult to inculcate these conclusions into clinical practice.
Also, by analyzing all above stage I post-operative patients (who were kept on follow up in view of favorable histology, and were not accessible for the present study), it would have been possible to compare the outcomes of neck irradiation alone to neck dissection alone. Therefore, a large well-randomized study is needed before clinically applying the results on the patient population.

Conclusion
To conclude, in early stage patients of carcinoma tongue, along with the management of local primary site, the neck should also be addressed, as the outcomes are significantly improved after primary surgical resection with concomitant neck dissection. 3,4,6 And based on our study, we conclude that elective neck irradiation can be considered as an alternative to neck dissection plus radiation to treat occult lymph node metastasis in stage I carcinoma oral tongue, but the same does not hold true for stage II patients in which neck dissection is mandatory.