Tumor control and functional outcome after radical surgery or interstitial radiotherapy-treatment options in T3/T4 cancer of the base of the tongue

Abstract

Purpose/Objective: The treatment modalities of preference for advanced T3/T4 squamous cell carcinoma of the base of tongue (BOT) in the Free University MC (Amsterdam) has been radical surgery with postoperative external beam radiotherapy (EBRT). The Erasmus MC (Rotterdam) treatment strategy for a similar subset of patients is focused on organ preservation. That is, EBRT to the primary and bilateral neck, followed by interstitial brachytherapy (BT) of the BOT, with neck dissection (ND) in case of N+ disease. This study compares local control (LC), survival, and functional outcome assessment of both treatment strategies. Materials/Methods: In FUMC (Amsterdam), between 1980–1996, radical surgery in combination with 50–60 Gy postoperative EBRT was performed in 58 patients with a median age of 58 years. The majority were men (67%). Two-thirds of patients had a T3 tumor. In the EMC (Rotterdam), brachytherapy (BT) for boosting primary cancers in the oropharynx was implemented routinely as of 1985. Thirty patients were analyzed in this retrospective review. Their median age was 58. Eighty-three percent were men. Two thirds of the patients had a T4 tumor. After a 1st series of EBRT to the primary and bilateral neck, 16 BOT tumors were subsequently implanted and boosted by LDR (1985–1991; median dose 28 Gy) and 14 by fractionated HDR (1992–1996; median dose 24 Gy). For BOT implants a quality index (Lowell Anderson) was computed. A univariate analysis was performed. Actuarial local (regional) control and survival, using Kaplan Meyer statistics, were computed. To analyze the functional performance of both treatment strategies, a questionnaire modified after List et al. (Cancer 1990; 66, 564-9) was distributed amongst all long-term survivors; 15 patients of the FUMC, and 13 from the EMC. For all long-term survivors, median scores of a visual analogue scale (VAS) for xerostomia were determined. Results: The distributions of sex, age and involvement of lymph nodes between the groups treated with primarily surgery and primary radiotherapy were comparable. The difference in distribution of tumor stage was statistically significant (p < 0.01). Four patients treated with primary radiotherapy had residual disease after completion of treatment. No significant difference was observed for the LDR- and fr.HDR brachytherapy schedules in terms of local (regional) control, and survival. The cumulative frequency of local failure at 3-years was 9% (FUMC) vs. 37% (EMC)(logrank 12.58, p < 0.01). If residual disease patients are excluded: 9% vs. 26%. The overall survival using log rank analysis did not show a significant difference (median 2.8 years for the Amsterdam vs. 2.5 years for the Rotterdam patients; (logrank 0.52, p = 0.47). The Lowell Anderson Q.I. was found to be non-discriminatory for LC and side-effects. The performance status of long term survivors revealed that the surgery patients fared less well; eating in public (53% vs. 77%, p = 0.19), normalcy of diet (69% vs. 87%, p = 0.26) and understandability of speech (73% vs. 92%, p = 0.19). The median VAS scores for xerostomia were 4.0 (FUMC) vs. 4.7 (EMC). Conclusions: The gender and age distribution of both series of patients were comparable. Given that 67% of the surgically treated patients had a T3 tumor, while 67% of those treated primarily radiation were categorized as T4, this could explain the lower LC rate of the EMC patients. Moreover, the OS curves of the FUMC and EMC patients are practically superimposible, with 40% surviving at 5-years. Also, the EMC patients fared better in functional outcome, that is in terms of eating in public, normalcy of diet and understandability of speech. In fact, the data presented illustrate once more that the QoL assessment is essential in determining the outcome when comparing clinically different treatment strategies. We feel the jury is still out on the optimal treatment of T3/T4 BOT cancer.