Abstract
Aim: To investigate the importance of chemotherapy and radiation dose after R1 resection of squamous cell carcinoma of the head-and-neck (SCCHN). Patients and Methods: One hundred and twenty-two patients receiving radiotherapy alone or with concurrent chemotherapy [cisplatin or cisplatin/5-fluorouracil (5-FU)] were retrospectively analyzed. Results: On multivariate analysis, chemotherapy was significantly associated with improved locoregional control (p=0.048). Three-year locoregional control rates were 61% for those treated without chemotherapy, 83% for those treated with cisplatin and 77% for those treated with cisplatin/5-FU. Radiation doses of 66 and 70 Gy were non-significantly superior to 60-64 Gy (p=0.18). On multivariate analysis, chemotherapy showed a trend for improving survival (p=0.055). Three-year OS rates were 51% for those without chemotherapy, 65% for those treated with cisplatin and 57% for those treated with cisplatin/5-FU. Radiation doses of 66 Gy (3-year survival=61%) and 70 Gy (70%) were superior to 60-64 Gy (25%) (p=0.021). Conclusion: Concurrent chemotherapy and a radiation dose of 66 Gy resulted in better outcomes. Cisplatin and cisplatin/5-FU were similarly effective. Radiation doses >66 Gy appear not to be necessary.
- Head-and-neck cancer
- microscopically incomplete resection
- palliative radiation therapy
- radiation dose
- completion of treatment
- survival
Patients with locally advanced (stage III/IV) squamous cell carcinoma of the head and neck region (SCCHN) have a comparably poor prognosis (1). Many patients undergo resection of the primary tumor and regional lymph nodes followed by radiotherapy or radiochemotherapy (2). Microscopically incomplete resection (R1 resection) is considered a risk factor for worse treatment outcomes (3). A re-analysis of two randomized trials suggested that patients in whom an R1 resection was performed benefited from the addition of chemotherapy to postoperative irradiation (4). Despite these randomized trials, two questions remain unanswered. One question relates to the chemotherapy regimen. Both randomized trials used an aggressive regimen consisting of 100 mg/m2 cisplatin alone given on days 1, 22 and 43 during the radiation course (5, 6). However, many centers worldwide use 5-fluorouracil (5-FU) in addition to cisplatin. The question is whether patients undergoing R1 resection would also benefit from cisplatin/5-FU. Another question relates to the most appropriate radiation dose. In one of the two randomized trials, all patients received 66 Gy (4), and in the other trial, 60 Gy or 66 Gy (5). Some centers also use 70 Gy. The present study investigated chemotherapy and compared no chemotherapy, cisplatin alone and cisplatin plus 5-FU. Additionally, three radiation dose levels were compared with respect to locoregional control and survival after R1 resection of non-metastatic stage III/IV SCCHN.
Patients and Methods
Data of 122 patients who received radiotherapy alone (n=45) or in combination with concurrent cisplatin-based chemotherapy (n=77) following R1 resection of locally advanced SCCHN were retrospectively analyzed for locoregional control and survival. Patients received radiotherapy of 2.0 Gy once daily on five days per week. Total doses ranged from 60-70 Gy (median dose=66 Gy) to the R1-resected former primary tumor. Higher-risk and intermediate-risk lymph nodes received 60 Gy and 50 Gy, respectively. When concurrent chemotherapy was administered, it included cisplatin alone (30-40 mg/m2 weekly, 20 mg/m2 on days 1-5+29-33, or 100 mg/m2 on days 1, 22 and 43) or cisplatin (20 mg/m2 on days 1-5+29-33) plus 600/1000 mg/m2 5-FU on days 1-5 and 29-33.
The impact of the following potential prognostic factors on locoregional control and survival was evaluated: Age (≤60 vs. >60 years), gender, pre-radiotherapy Karnofsky performance score (80-100 vs. ≤70), site of SCCHN (oropharynx vs. hypopharynx vs. larynx vs. oral cavity/floor of mouth), T-stage (T1/T2 vs. T3/T4), N-stage (N0/N1 vs. N2/N3), histological grade (G1/G2 vs. G3), pre-radiotherapy hemoglobin (<12 vs. ≤12 g/dl), radiotherapy dose (60-64 Gy vs. 66 Gy vs. 70 Gy) and concurrent chemotherapy (none vs. cisplatin alone vs. cisplatin/5-FU).
The univariate analyses of locoregional control and survival were performed using the Kaplan–Meier analysis supplemented by the log-rank test. Potential prognostic factors achieving significance (p<0.05) or a trend (p<0.08) on univariate analyses were additionally evaluated in a multivariate manner with the Cox regression model.
Results
Patients were followed up until death or for a median of 28 months (range=5-76 months) in those alive at the last follow-up. For the entire cohort, the locoregional control rates at 3 and 4 years were 73% and 66%, respectively. Of all investigated factors, only chemotherapy was found to have a significant association with locoregional control (Table I). The 3- and 4-year rates were 61% and 33%, respectively, for patients who did not receive chemotherapy; 83% and 83% respectively, after radiochemotherapy with cisplatin alone; and 77% and 77%, respectively, after radiochemotherapy with cisplatin/5-FU (p=0.010, Figure 1). On the Cox regression analysis, the addition of chemotherapy to radiotherapy was also significant (risk ratio=1.64, 95% confidence interval=1.01-2.77; p=0.048).
The 3- and 4-year survival rates for the whole patient cohort were 58% and 54%, respectively. On univariate analyses, lower N-stage (p=0.006), lower histological grade (i.e. better differentiated tumors) (p=0.021) and pre-radiotherapy hemoglobin levels of ≥12 g/dl (p=0.002) were significantly associated with improved survival (Table II). In addition, lower T-stage (p=0.053) and RT doses of 66 Gy or 70 Gy (p=0.076) showed a trend for association with better survival. These five factors plus chemotherapy were included in the Cox regression analysis. In that analysis, T-stage (p=0.018), N-stage (p=0.011), pre-radiotherapy hemoglobin level (p=0.003) and the radiotherapy dose (p=0.021) achieved significance. Histological grade (p=0.078) and chemotherapy (p=0.055) showed a trend. The complete results of the Cox regression analysis are summarized in Table III.
Locoregional control rates at 3 and 4 years (univariate analysis).
Discussion
Patients with metastatic SCCHN have a very poor prognosis (7, 8). Patients with locally advanced non-metastatic disease have a better expected outcome (1). In many patients with resectable stage III/IV SCCHN, the final pathological evaluation reveals that the tumor has not been removed completely microscopically. The question is whether a second surgery should be performed. A re-resection is often not possible or refused by patients. There is general agreement that patients undergoing resection of stage III/IV tumors should receive postoperative radiotherapy (3, 9). Since two randomized trials and their re-analysis demonstrated 10 years ago that patients in whom only R1 resection was performed benefited from the addition of concurrent chemotherapy, this approach became the standard procedure in many centers worldwide (4-6). However, radiochemotherapy after R1 resection is not performed everywhere. When concurrent chemotherapy is administered, the most appropriate chemotherapy protocol is still controversial. In the two randomized trials, an aggressive protocol with three courses of 100 mg/m2 of cisplatin was used. This regimen is the standard in many countries, including the United States. Since this regimen is fairly toxic and the patients require substantial supportive care, other centers, particularly in Europe, have explored alternatives. One alternative is a regimen including two courses of 20 mg/m2 cisplatin given for five days in the first and fifth week of radiotherapy (10). This regimen has been shown to be much better tolerated than three courses of 100 mg/m2 of cisplatin (11). However, since the cumulative cisplatin dose of two courses of 20 mg/m2 cisplatin is only two-thirds that of three times 100 mg/m2, many centers added two courses of 5-FU (either 600 or 1,000 mg/m2) over five days in the first and fifth week of radiotherapy to cisplatin at 20 mg/m2. To our knowledge, it has not yet been shown that after R1 resection, patients benefit from radiochemotherapy with cisplatin/5-FU.
Locoregional control of the groups with no chemotherapy treated with cisplatin alone and treated with cisplatin/5-fluorouracil (5-FU).
In this study, radiochemotherapy either with cisplatin alone or cisplatin/5-FU was superior to radiotherapy alone. Cisplatin/5-FU resulted in similar locoregional control and survival rates as different cisplatin alone regimens. Moreover, the outcomes were similar to those found in the two previous randomized trials using high-dose cisplatin (three courses of 100 mg/m2). In order to better define the optimal chemotherapy regimen for radiochemotherapy of R1-resected SCCHN, additional studies directly comparing these regimens are warranted. Novel approaches including induction chemotherapy and 5-FU pro-drugs may be reasonable (12, 13).
Survival rates at 3 and 4 years (univariate analysis).
Another important question addresses the most appropriate total radiation dose. One of the previous randomized trials used 66 Gy for all patients, the other trial either 60 Gy or 66 Gy. In the latter trial, a separate analysis of the separate dose groups was not performed. Thus, the optimal dose is unclear. Is 60 Gy sufficient or should the dose be 66 Gy or even greater? In the present study, 66 Gy was clearly superior to 60-64 Gy, as was a dose of 70 Gy. However, an escalation of the radiation dose beyond 66 Gy did not appear to further improve outcomes significantly. Thus, 66 Gy appears to be appropriate.
In addition to concurrent chemotherapy and the radiation dose, T-stage, N-stage and pre-radiotherapy hemoglobin were significantly associated with survival. Histological grade showed a trend towards such an association. These findings agree with previously reported data, which reveals some consistency of the results of the present study (14-17). However, its retrospective nature should be taken into account when interpreting the data. Prospective randomized trials focusing on R1-resected SCCHN are desirable but may not be practical, since R1 resection is less common due to improved surgical procedures and standards.
Multivariate analysis (Cox regression model) of survival.
In conclusion, this study supported the importance of concurrent chemotherapy when added to radiotherapy for R1-resected SCCHN. Cisplatin alone and cisplatin/5-FU appeared to be similarly efficacious. The optimal chemotherapy regimen needs to be defined. 66 Gy appears to be an appropriate radiation dose.
Footnotes
Conflicts of Interest
On behalf of all Authors, the corresponding Author states that there are no conflicts of interest related to this study.
- Received February 24, 2016.
- Revision received April 1, 2016.
- Accepted April 6, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- Predicting the Ambulatory Status of Patients Irradiated for Metastatic Spinal Cord Compression (MSCC) from Head-and-neck Cancer
- A Tool to Predict the Probability of Intracerebral Recurrence or New Cerebral Metastases After Whole-brain Irradiation in Patients with Head-and-Neck Cancer
- Hyperfractionated or Accelerated Hyperfractionated Re-irradiation with >=42 Gy in Combination with Paclitaxel for Secondary/Recurrent Head-and-Neck Cancer