Abstract
Background/Aim: Locoregional recurrences of head and neck cancers are often associated with a poor overall survival (OS). Outcomes may be improved with individualized treatments considering a patient's lifespan. A specific scoring system for estimating OS prognoses is presented. Patients and Methods: In a preceding study of these 60 patients, Eastern Cooperative Oncology Group performance score (ECOG 0-1 versus 2, p=0.002) and N-stage (N0-1 versus N2, p=0.004) were identified as independent predictors of OS. Results: In the current study, the following scoring points were assigned: ECOG performance score 2=0 points, ECOG performance score 0-1=1 point, N2=0 points, N0-1=1 point. Patient scores were 0, 1 or 2 points with 3-year OS rates of 0%, 46% and 83%, respectively (p<0.0001). On multivariable analysis, differences remained significant (p<0.0001). Conclusion: This new scoring system includes three groups of patients with significantly different OS prognoses and can assist physicians when designing individualized therapy for locoregional recurrences of head and neck cancer.
Locoregional recurrence is quite common in patients treated for head and neck cancer and is often associated with a poor overall survival (OS) (1, 2). In addition to modern technology including intensity-modulated radiotherapy (IMRT) or volumetric-modulated arc therapy (VMAT), the outcomes of these patients may be improved with the administration of individualized treatment approaches. Individualized therapy should account for the patient's personal preference, social situation, age, co-morbidity index, and remaining lifespan. In patients with short expected OS, short and less burdensome treatment should be preferred, whereas in patients with a comparably long remaining lifetime, long-term locoregional control and reduction of the risk of developing late treatment-related morbidity are more important (3-7). These considerations lead to the conclusion that it is important to be able to estimate an individual patient's OS time. In a previous study, we identified independent predictors of OS in a series of patients irradiated for a locoregional recurrence of head and neck cancer (8). However, a specific scoring system would be more convenient and easier to use during daily clinical care. Therefore, the current study was conducted to create a specific scoring system to estimate the survival of patients irradiated for locally recurrent head and neck cancer.
Patients and Methods
A total of 60 patients were irradiated for locoregional recurrence of squamous cell carcinoma in the head and neck region and included in this retrospective analysis. In a preceding study, 12 potential prognostic factors were investigated for a potential impact on OS following radiotherapy. These factors were age, gender, primary tumor site, initial tumor stage, type of primary treatment, time from first diagnosis of the head and neck tumor until the beginning of radiation therapy for the recurrence, performance status, T-stage of the recurrence, N-stage of the recurrence, upfront surgery, simultaneous chemotherapy and the total dose of radiotherapy. Of these variables, the Eastern Cooperative Oncology Group performance score (ECOG 0-1 versus ECOG 2; hazard ratio=3.69; p=0.002) and N-stage (N0-1 versus N2; hazard ratio=3.20; p=0.004) were significantly (defined as p<0.05) associated with OS on multivariable analysis in our previous publication (8). Therefore, both ECOG performance score and N-stage were included in the scoring system created in the present study.
Results
Median follow-up times were 17.5 months (range: 1-96 months) for the entire cohort and 36 months (range: 6-96 months) for those patients who were still alive at their last follow-up. The following points were assigned to the variables: ECOG performance score 2=0 points, ECOG performance score 0-1=1 point, N2=0 points, N0-1=1 point. The points were summed for individual patients and were 0, 1 or 2 points. The OS rates at 1, 2 and 3 years associated with the different scores are summarized in Table I. The three possible scores were compared for OS using the Kaplan–Meier method supplemented by the log-rank test (Figure 1), which demonstrated that the difference between the three total scores was significant (p<0.0001). The median OS times were 5 months, 26 months and >39 months (median not reached), respectively. The total score was an independent predictor of OS as confirmed by multivariate analysis using the Cox proportional hazards model. Indeed, the impact of the total score on OS remained significant (hazard ratio=3.35, 95%-confidence interval=1.89-6.02, p<0.0001).
Discussion
The outcomes of patients with loco-regional recurrences of head and neck cancer are poor and may be improved with the strategy of personalized treatment (9). As in other oncological situations, a patient's expected OS is an important consideration when choosing the best treatment regimen (3-7). In a previous study of 60 patients receiving radiotherapy for locoregional recurrence of head and neck cancer, we identified two independent prognostic factors independently associated with OS (8). Such prognostic factors play an important role when physicians tailor their treatment regimen to a specific individual. In order to facilitate personalized care, we created a scoring system from that cohort that allows estimation of the OS prognosis of patients. This scoring system was based on the two previously identified independent prognostic factors, namely ECOG performance score and N-stage at recurrence (8). Three prognostic groups were identified, i.e. those with 0 points, 1 point and 2 points. According to the results of the present study, these groups were associated with significantly different OS prognoses (Table I).
In the group with 0 points, only 18% of the patients survived for 1 year following radiotherapy, and the median OS time was only 5 months. Therefore, these patients may be considered as requiring palliative therapy and should probably receive a radiation regimen that is not aggressive or burdensome. Furthermore, the overall treatment time should be kept as short as possible. Shorter hypofractionated regimens such as 13×3 Gy in 2.5 weeks or 18×2.5 Gy in 3.5 weeks may be more appropriate options than conventionally fractionated radiotherapy with 2-Gy fractions up to 60-70 Gy over 6 to 7 weeks. Patients of the group with 1 point had an intermediate OS prognosis, with 50% surviving for at least 2 years following radiotherapy and a median OS time of 26 months. Therefore, these patients may not be ideal candidates for a palliative approach with hypofractionated radiotherapy and should receive conventional radiotherapy over 6-7 weeks instead. If complete resection is not performed or patients with head and neck cancer have an extracapsular extension of lymph node metastases, radiotherapy should be administered with concurrent chemotherapy. If no contraindications exist, chemotherapy should be cisplatin-based (10-13). Other options of systemic therapies include carboplatin, mitomycin C plus 5-fluorouracil, cetuximab or paclitaxel (11, 14-16). Those patients who achieved 2 points had the most favorable OS prognoses, with a 3-year OS rate of 83%. Patients within this group would likely benefit from the same recommendations as the group with 1 point. However, because the risk of radiation-related late morbidity increases over time, treatment for these more favorable patients in the group with 2 points should include greater efforts at sparing of normal tissues such as parotid glands, larynx, swallowing muscles and others in order to prevent late toxicity. Therefore, these patients should be irradiated with modern techniques such as IMRT and VMAT. When interpreting these results, one should bear in mind that the scoring system was developed from retrospective data. Therefore, inclusion of a hidden selection bias might exist.
Kaplan–Meier curves of the three prognostic groups (0, 1 and 2 points) for overall survival (OS). The p-value was calculated with the log-rank test.
Overall survival (OS) following radiotherapy of a locoregional recurrence of head and neck cancer.
In summary, this new scoring system divides patients irradiated for recurrent head and neck malignancies into three groups with significantly different OS prognoses. This score provides a helpful tool to assist physicians in estimating survival and choosing an individualized treatment approach for patients with locoregional recurrence of head and neck cancer. It could also be used for patient stratification in future trials designed for patients with recurrent disease.
Footnotes
Conflicts of interest
On behalf of all Authors, the corresponding Author states that there is no conflict of interest related to this study.
- Received December 4, 2017.
- Revision received January 4, 2018.
- Accepted January 10, 2018.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.