Adjuvant Therapy for Advanced Head and Neck Cancer: More is Better
ABSTRACT & COMMENTARY
Synopsis: Postoperative radiotherapy with concurrent cisplatin leads to improved local control for patients with resectable, locally advanced squamous cell cancers of the head and neck, which exhibit features indicative of a higher risk of recurrence (e.g., extracapsular nodal extension).
Source: Al-Sarraf M, et al. Int J Radiat Oncol Biol Phys 1997;37:777-782.
Standard therapy for advanced, non-metastatic head and neck cancer consists of an en bloc surgical resection followed by postoperative radiotherapy. The identification of adverse risk factors, predictive of local-regional failure and/or distant metastatic disease, permits the development and testing of more aggressive therapies in those selected populations. Nguyen et al1 retrospectively studied the pattern of relapse and prognostic factors for patients with advanced laryngeal cancer treated with postoperative radiotherapy between 1980-1985. Higher doses of radiation were used for patients with close or positive margins, residual disease, and/or extracapsular nodal extension (ECE) of tumor. Of 116 patients treated, 12 had positive margins, 23 had residual tumor, and 45 had extracapsular extension of tumor. With a minimum of three years and a median of five years of follow-up, 27 patients have failed locally (24 local only with 23 deaths), and 18 relapsed with distant disease, 15 with distant relapse only. Of the many prognostic factors studied (e.g. primary site, T stage, N stage, and surgical margins), only extracapsular extension of tumor identified patients who had a greater risk of local failure and death. The local recurrence rates observed in this study were similar for patients who had negative margins, positive margins, or residual tumor after surgery, suggesting that the delivery of higher dose RT to high-risk patients improves local control. However, there was still a local recurrence rate of more than 20%.
The use of cisplatin-based chemotherapy before surgery, between surgery and RT, or following RT has led to some decrease in systemic metastases but hasn’t really changed the incidence of local-regional recurrence or improved overall survival. Concurrent radiochemotherapy, which has improved locoregional control, disease-free survival, and perhaps even overall survival compared to RT alone in patients with advanced unresectable head and neck cancer, was employed by Al-Sarraf and colleagues to improve local control in patients with advanced, but resectable, head and neck cancer. They reported the results of Radiation Therapy Oncology Group (RTOG) study 88-24, in which patients with Stage IV disease or histologic proof of involved surgical margins received cisplatin 100 mg/m2 IV to coincide with days 1, 22, and 43 of RT. RT was designed to deliver 60 Gy in six weeks with pathologically uninvolved areas of the neck receiving at least 45 Gy. Ninety percent of patients received at least 90% of their protocol-prescribed RT. Only 61% of patients received all three cycles of cisplatin, and an additional 25% received two cycles. The most common reasons for not receiving cisplatin were toxicity or patient refusal. Distant metastases were the most common cause of treatment failure (70% of all failures), whereas the local failure rate was a very low 19% at three years. This represented a significant improvement in local control compared to historical controls culled from a previous Intergroup study of patients treated with RT alone. Al-Sarraf et al estimate that concurrent cisplatin resulted in a 55% reduction in risk of locoregional relapse. Unfortunately, there was no apparent effect on survival.
The results of this single-arm study are further supported by the data from Bachaud and associates,2 who reported on the final results of a randomized trial in which patients with histological evidence of ECE of disease were treated with postoperative radiotherapy and randomized to receive concurrent weekly cisplatin (50 mg) for 7-9 weeks or no additional treatment. The radiation dose was 54 Gy to the primary site and draining lymphatics with a boost to the primary site for positive margins, and a boost to the nodal site of ECE to 65-74 Gy using electron beam fields. A total of 88 patients were randomized between 1984 and 1988; 44 patients in the RT arm and 39 patients in the CMT group were evaluable. Local failure was observed in 41% of patients in the RT arm compared to 23% in patients treated with CMT (P = 0.08). The incidence of distant failure was similar in both groups, 30% and 26%, respectively. The two- and five-year survival rates were 46% and 13% in the RT group and 72% and 36% in the CMT group, respectively. Serious late toxicities were not seen more frequently in the patients treated more aggressively, but no comment was made regarding the acute morbidity and mortality of treatment.
Patients with locally advanced head and neck cancer are more prone to die from local regional relapse than from distant metastatic disease. This has long been the rationale for the addition of postoperative radiotherapy. Since RT alone has a significant failure rate, investigators have searched for effective strategies to integrate chemotherapy. The hope is that improvements in local control and decreases in the number of distant relapses will be seen following chemotherapy. The increased risk of recurrence for patients with ECE of tumor reported by Nguyen et al1 confirm those from a series using postoperative radiotherapy from the MD Anderson Cancer Center, where patients with ECE had an actuarial risk of 30% for isolated local regional relapse. These patients will benefit from more aggressive treatment. The higher doses of RT used in the trial reported above were those that are maximally tolerated by the tissues treated. Higher doses cannot be delivered with conventional RT, but these doses can be delivered with similar complication rates with hyperfractionated therapy; however, this has not been tested in a randomized trial for postoperative adjuvant therapy.
Systemic chemotherapy for head and neck cancer has been used mostly as neoadjuvant treatment for locally advanced disease or for palliation of metastatic or recurrent disease. While impressive local responses have been observed following neoadjuvant therapy, no improvement in local, regional control, or survival has been detected. Al-Sarraf et al report impressive local control rates with high doses of cisplatin given concurrently with radiation. The purported benefit of cisplatin is found when comparing the current results with historical controls, an approach fraught with many problems. Toxicity is substantial, as severe and life-threatening toxicities occurred in 20% and 12% of patients, respectively.
The trial by Bachaud et al2 was randomized and found that for patients with the highest risk of recurrence, there was a benefit from the addition of weekly concurrent cisplatin. The small size of the trial and the variety of primary sites that were represented in the patient population make their conclusions somewhat tenuous. However, similar results showing a benefit with concurrent chemotherapy in the postoperative setting have also been reported using mitomycin-C.3 Combined modality therapy is tolerable albeit with an increase in acute toxicities, but no increase in long-term side effects of treatment has been reported. Both cisplatin and mitomycin-C are effective when combined with radiotherapy, cisplatin for the treatment of lung and bladder cancer, and mitomycin-C for anal cancer. Because of the increase in local toxicities from combined modality therapy, it is important to limit the use of this approach to those patients that require more aggressive treatment. The finding of extracapsular extension of disease for patients with head and neck cancer should alert the clinician to the need for a combined modality approach, preferably with either cisplatin or mitomycin-C. A randomized Intergroup trial is also available for these patients to verify the benefit from concurrent cisplatin therapy. It is likely that we will find similar improvements in local control in this study, which will establish combined modality therapy as the standard for patients with advanced, resectable head and neck cancer with negative prognostic features. It is likely that the mechanism of this activity is the radiosensitizing properties of the chemotherapy. The failure to document decreases in rates of distant metastases or to improve survival, argues against chemotherapy killing of residual tumor cells as the explanation. Although improved local control is a laudable goal and a significant achievement, major decreases in the death rate from head and neck cancer await changes in lifestyle, or the development of effective chemoprevention strategies and/or better systemic therapy.
1. Nguyen TD, et al. Int J Radiat Oncol Biol Phys 1996; 36:1013-1018.
2 Bachaud JM, et al. Int J Radiat Oncol Biol Phys 1996; 36:999-1004.
3. Hafty BG, et al. Int J Radiat Oncol Biol Phys 1993;27: 241-250.