Japanese Journal of Clinical Oncology

INTRODUCTION

Patients with locoregionally advanced nasopharyngeal cancer generally have a high incidence of both locoregional and distant failure, despite recent technical progress in imaging and radiation therapy (1). Therefore, strategies which improve locoregional control and eradicate occult systemic disease should be encouraged.

Although the radiation dose-response relationship has not been clearly demonstrated for nasopharyngeal cancer (2,3), a larger total dose with adequate coverage and suitable technique might improve locoregional control. To deliver higher radiation doses without increasing normal tissue damage, hyperfractionated radiotherapy would appear to be beneficial.

Nasopharyngeal cancer is known to be highly sensitive to chemotherapy. To improve both locoregional and distant control, additional use of systemic chemotherapy in a neoadjuvant or concurrent setting has been tried. Despite the excellent initial response, the value of neoadjuvant chemotherapy (NAC) on long-term locoregional control in nasopharyngeal cancer is still controversial (4-8). Al-Sarraf et al. (9) reported promising results using concurrent chemotherapy and radiation therapy. However, they reported a high incidence of severe acute toxicities with concurrent treatment. Severe toxicity, especially mucositis, might be due to large irradiation portals, which include large areas of mucous membrane in the case of nasopharyngeal cancer. To eradicate systemic microscopic disease, adjuvant chemotherapy (ACT) was considered to be suitable. We thought that ACT would avoid severe mucosal toxicity compared with concurrent chemotherapy.

Taking into account this background, we had carried out a pilot study of hyperfractionated radiotherapy and ACT on patients with locoregionally advanced nasopharyngeal cancer. In this paper, we review our experience of this strategy on seven patients.

PATIENTS AND METHODS

Seven patients with previously untreated histologically proven nasopharyngeal cancer treated with hyperfractionated radiotherapy and adjuvant chemotherapy were reviewed. Table 1 shows the patients' characteristics. Patients were re-staged according to the 1997 UICC classification. Case 2 was staged stage IV (T2N2bM0) according to the 1987 UICC classification at the time of treatment. One patient (No. 5) had received radical neck dissection and open biopsy in sphenoid sinus before entry to our protocol.

Table 1. Summary of patients' characteristics

No.	Age/gender	PS	T*	Skull invasion	Cranial nerve impairment	N*	Stage*	Pathology (WHO)	LDH
1	54M	0	3	+	-	2	3	2	417
2	59M	0	1	-	-	1	2b	1	473
3	43M	1	4	+	+	2	4a	2	315
4	33M	1	4	+	+	2	4a	2	324
5	52M	1	4	+	+	2	4a	2	292
6	54M	0	3	+	-	0	3	2	394
7	63M	0	3	+	-	2	3	1	706

*UICC 1997 classification.

Table 2. Acute toxicity

No.	Hyperfractionated radiotherapy				Adjuvant chemotherapy (WHO grade)						BW loss (%)
	RTOG grade		Nutrition	RT break/day	N/V	WBC	Plt.	Mucosa	Renal	Hepato
	Mucosa	Skin
1	3	1	Purée diet	0	2	3	1	1	0	0	15.4
2	3	1	Soft diet	5 (holiday)	0	2	0	0	0	0	12.4
3	3	2	Liquid diet	0	1	2	0	2	0	0	7
4	3	1	Soft diet	10	2	2	0	2	0	0	1.9
5	3	2	IVH	0	2	2	2	1	0	0	10.8
6	3	1	Soft diet	0	1	3	2	2	0	0	7.7
7	3	2	Soft diet	0	2	2	1	0	0	0	3.7

N/V, nausea and vomiting.

Pre-entry work-up required inspection of the nasopharynx, palpation of the neck, CT scans and MRI of the head and neck (including nasopharynx, skull base and cervical nodes), chest X-rays, bone scintigraphy and liver ultrasound.

The primary tumor and the upper neck were treated with parallel opposed lateral portals and the lower neck was irradiated bilaterally with parallel opposed anteroposterior portals. A dose of 1.2 Gy was administered twice a day at each treatment session through the upper neck portals. The minimum interfraction interval was 6 h. The spinal cord was shielded after 40.8 Gy/34 fractions. After this, 9 or 12 MeV electron beams were used to treat posterior neck. The lower neck was treated once a day with a dose of 2 Gy. The subclinical disease sites were treated to a total dose of 50 Gy. The total dose to the primary tumor was 76.8Gy/64 fractions except for one patient (No. 5). He received 81.6Gy/68 fractions. Initial large portals were treated with a 4 MV photon beam and boost portals to the primary tumor with a 10 MV photon beam. Involved lymph nodes were boosted to a total dose of 60-70 Gy using 9 or 12 MeV electron beams. The photon beam dose was calculated at the mid-depth of the central axis. No dose compensators were used.

ACT consisted of cisplatin at 80 mg/m² i.v., infused over 2 h on day 1 with hyperhydration and 5-FU at 800 mg/m² on days 2-6, c.i. (120 h). All patients received antiemetics with granisetron at 3 mg and metcropramide prior to the cisplatin infusion. This schedule was repeated every third week for two cycles.

Acute radiation morbidity scoring criteria developed by the Radiation Therapy Oncology Group (RTOG) was employed to assess acute toxicities resulting from hyperfractionated radiotherapy (10). For toxicity assessment on the mucous membrane, patches (<3 mm) were defined as grade 2 and confluent mucositis ([ge]3 mm) was defined as grade 3.

Toxicities resulting from ACT were recorded according to the WHO recommended criteria (11).

RESULTS

Table 3. Treatment outcome

No.	Site of failure	Complication	Status
1	-	-	48 m NED
2	-	-	41 m NED
3	Bone, liver	Soft tissue necrosis	35 m AWD
4	Primary	-	33 m DWD
5	Lung	Brain necrosis (focal)	28 m AWD
6	-	-	25 m NED
7	Primary	-	25 m AWD

NED, no evidence of disease; AWD, alive with disease; DWD, dead with disease; m, months.

DISCUSSION

Some schedules of altered fractionated radiotherapy have been investigated to improve locoregional control for locally advanced nasopharyngeal cancer (8,12,13). Most studies employed an accelerated hyperfractionation schedule which mainly emphasized a shortening of total treatment time. In our study, we selected the hyperfractionated radiotherapy schedule (1.2 Gy twice daily) (14), which mainly emphasized increasing total radiation dose. In our analysis, five of the seven patients were locoregionally controlled. Despite the very small number of patients included in this series, we consider this an encouraging result. Six of the seven patients included in this study would be staged T4 using the 1987 UICC classification. In previous publications, the incidence of local failure for T4 patients treated with radiotherapy has been reported to be [sim]30-100% (15). On the other hand, although acute toxicities were almost acceptable, two patients suffered severe late complications. This suggests that hyperfractionated radiotherapy with 76.8-81.6 Gy carries some risk of late complications. Cmelak et al. (16) reported promising results using radiosurgery in the treatment of locally advanced nasopharyngeal cancer. Higher doses might be safely delivered to residual primary tumors with three-dimensional (3-D) treatment in conformal therapy.

Some reports are available concerning the value of ACT for nasopharyngeal cancer (17,18). Although Rossi et al. (17) found no significant value of ACT in their randomized study, their regimen did not contain cisplatin. In our series, two patients developed distant metastases despite limited follow-up periods. This suggested that two cycles of chemotherapy using a relatively standard dose of cisplatin and 5-FU might not be very effective in eradicating micrometastases.

In summary, these results suggest that our regimen of hyperfractionated radiotherapy and adjuvant chemotherapy was almost well tolerated and might be of use in locoregional control for nasopharyngeal cancer. However, it carries some risk of late complications. The 3-D conformal irradiation technique and adequate dose intensity chemotherapy might be encouraged to improve the outcome of patients with locoregionally advanced nasopharyngeal cancer.