Abstract
Aim: To identify predictors of locoregional control (LRC) and overall survival (OS) after definitive radio(chemo)therapy for squamous cell carcinoma of the head and neck (SCCHN). Patients and Methods: Two hundred and seventy-five patients were evaluated; 261 patients received radiochemotherapy with 30-40 mg/m2 of cisplatin weekly, three courses of cisplatin 100 mg/m2, two courses of cisplatin 5x20 mg/m2 or two courses of cisplatin 5×20 mg/m2plus 5-fluorouracil. Ten characteristics were analyzed: Pre-radiotherapy hemoglobin, T-/N-category, Karnofsky performance-score (KPS), gender, age, chemotherapy type, tumor site, grading and radiation dose. Results: On multivariate analyses, hemoglobin 12-14 g/dl (p=0.040), lower T-category (p=0.010), lower N-category (p=0.042) and female gender (p=0.006) were predictive of LRC. Hemoglobin >12 g/dl (p=0.020), lower N-category (p<0.001), KPS ≥80 (p<0.001), female gender (p=0.024) and cisplatin 100 mg/m2 or 5×20 mg/m2 (p<0.001) were predictors of improved OS. Conclusion: Predictors of LRC and OS were identified that can improve personalization of treatment. Since chemotherapy type was associated with OS, studies comparing different regimens are warranted.
- Locally advanced SCCHN
- definitive radio(chemo) therapy
- prognostic factors
- locoregional control
- overall survival
The most common treatments for locally advanced squamous cell carcinoma of the head and neck (SCCHN) include surgical resection followed by radio(chemo)therapy and definitive radio(chemo)therapy (1). The preference of treatment approaches depends on local expertise, national and institutional standards. Often, patients receiving definitive radio(chemo)therapy have large unresectable primary tumors, extensive lymphadenopathy, poorer performance and higher comorbidity than patients considered suitable for resection. Therefore, many patients treated with definitive radio(chemo)therapy for locally advanced SCCHN have a poor prognosis, requiring further research. In addition to modern radiation techniques and novel anticancer drugs, the use of more individualized treatment approaches can lead to better outcomes after definitive radio(chemo) therapy for patients with SCCHN (2-4). Besides age, comorbidity, personal situation and treatment preference, individualized therapeutic approaches should take into account the expected outcomes of treatment, such as locoregional control (LRC) and overall survival (OS). Both LRC and OS can be estimated with profound knowledge of independent prognostic factors. Furthermore, the most appropriate radiation dose and type of concurrent chemotherapy are critical to understand. This study analyzed the prognostic role of radiation dose, different dosing regimens of cisplatin-based chemotherapy and additional potential prognostic factors on LRC and OS in patients receiving definitive radio(chemo)therapy for locally advanced SCCHN.
Patients and Methods
Two hundred and seventy-five patients who received definitive radio(chemo)therapy for locally advanced SCCHN were evaluated for LRC and OS in this retrospective study. Irradiation was performed as conventionally fractionated (2.0 Gy per weekday) three-dimensional conformal radiotherapy. Total radiation doses administered to the primary tumor and the involved lymph nodes ranged between 60 Gy and 70 Gy. Doses given to non-involved intermediate-risk and high-risk lymph nodes were 50 Gy and 60 Gy, respectively. All but 14 patients received simultaneous chemotherapy in addition to radiotherapy, either with cisplatin weekly (30-40 mg/m2 each week), three courses of cisplatin 100 mg/m2 (days 1, 22 and 43), two courses of cisplatin 20 mg/m2 (days 1-5 and 29-33) or two courses of cisplatin 20 mg/m2 supplemented by 600/1,000 mg/m2 5-fluorouracil (5-FU) (days 1-5 and 29-33).
Ten characteristics were analyzed for potential associations with LRC and OS, including pre-radiotherapy hemoglobin level (<12 g/dl versus 12-14 g/dl vs. >14 g/dl), T-category (T1/T2 versus T3/T4), N-category (N0/N1/N2a vs. N2b/N2c/N3), Karnofsky performance-score (≤70 versus ≥80), gender, age (≤57 years versus ≥58 years, median age=57 years), type of chemotherapy (none versus cisplatin weekly versus cisplatin 100 mg/m2 versus cisplatin 20 mg/m2 versus cisplatin 20 mg/m2 plus 5-FU), primary tumor site (oropharynx versus hypopharynx versus larynx versus oral cavity/floor of mouth), histologic grading (G1/G2 versus G3) and total radiation dose (60-66 Gy versus 70 Gy).
The Kaplan-Meier method plus the log-rank test were applied for univariate analyses of LRC and OS (5). Characteristics showing a significant association (p<0.05) with LRC or OS were also included in a multivariate analysis performed with the Cox proportional hazards model.
Results
The LRC rates in the entire series were 73%, 60% and 58%, respectively, after 1, 3 and 5 years following irradiation. On univariate analyses, LRC was positively associated with a pre-radiotherapy hemoglobin level of 12-14 g/dl (p=0.003), lower T-category (p=0.029), lower N-category (p=0.020) and female gender (p=0.009). Results of the univariate analyses of LRC are shown in Table I. On the multivariate analysis, pre-radiotherapy hemoglobin level (p=0.040), T-category (p=0.010), N-category (p=0.042) and gender (p=0.006) proved to be independent predictors of LRC (Table II).
OS rates after 1, 3 and 5 years were 76%, 50% and 36%, respectively. In the univariate analyses, OS was positively associated with pre-radiotherapy hemoglobin levels of 12-14 g/dl and >14 g/dl (p=0.019), lower N-category (p<0.001), a KPS of ≥80 (p<0.001), female gender (p=0.045), chemotherapy with three courses of cisplatin 100 mg/m2 or two courses of cisplatin 20 mg/m2 (p=0.009) and larynx or oropharynx cancer (p<0.001). The results of the univariate analyses of OS are summarized in Table III. In the multivariate analysis of OS, pre-radiotherapy hemoglobin level (p=0.020), N-category (p<0.001), KPS (p<0.001), gender (p=0.024) and type of simultaneous chemotherapy (p<0.001) were significant predictors, whereas the primary tumor site did not achieve significance (p=0.57). The complete results of multivariate analyses of OS are shown in Table IV.
Discussion
The prognoses of patients with SCCHN presenting with distant metastases or locally advanced unresectable disease are generally poor and require improvement (1-3, 6-8). For locally advanced disease, modern techniques of head-and-neck surgery and radiotherapy have already led to improved outcomes, which may be further enhanced with new chemotherapy approaches (9, 10). The application of personalized treatment programs may also result in better outcomes. Such programs should consider an individual patient's specific situation, including his prognosis. Therefore, the identification of prognostic factors is important. In this study, several potential predictors for LRC and OS were analyzed. In the multivariate analysis of LRC, pre-radiotherapy hemoglobin levels of 12-14 g/dl, lower T-category, lower N-category and female gender were independent prognostic factors. In the multivariate analysis of OS, pre-radiotherapy hemoglobin levels of >12 g/dl, lower N-category, KPS ≥80, female gender and the type of the concurrent chemotherapy were independent predictors. These results widely agree with those of previous studies of treatment of SCCHN demonstrating consistency of the results of this study (11-13). An impact of the tumor stage on outcomes was already shown for patients with locally advanced SCCHN. The prognostic role of the pre-radiotherapy hemoglobin was also reported before, for which two interpretations exist. The hemoglobin level may be a surrogate marker for more advanced underlying disease. However, it has also been suggested that hemoglobin levels are important for tumor oxygenation, which is important for the efficacy of radiation therapy (14, 15). The effect of irradiation widely depends on the induction of oxygen free radicals that lead to damage of the tumor cell DNA and, subsequently, the tumor cell itself. It is well-recognized that anemia, which is associated with a reduced oxygen-carrier capacity, has a detrimental effect on tumor cell oxygenation (14). In addition, it has been suggested that high concentrations of hemoglobin may also have a negative impact on tumor cell oxygenation due to a reduced perfusion of the small capillaries (16). Hemoglobin levels of 12-14 g/dl or 13-15 g/dl have been considered optimal for tumor oxygenation. Indeed, in the present study, hemoglobin levels of 12-14 g/dl were associated with a better LRC than levels >14 g/dl and <12 g/dl. This result supports the idea of an optimal range of hemoglobin levels in the light of tumor cell oxygenation. With respect to OS, both hemoglobin levels of 12-14 g/dl and >14 g/dl were superior to levels <12 g/dl, while the OS rates for 12-14 g/dl and >14 g/dl were similar. Thus, in addition to the aspect of tumor cell oxygenation, the hemoglobin level may also be a surrogate marker for advanced disease.
In the present study, the radiochemotherapy protocols, including three courses of 100 mg/m2 cisplatin (on days 1, 22 and 43) or two courses of 20 mg/m2 cisplatin (on days 1-5 and 29-33), resulted in better OS than the protocols that included weekly administration of cisplatin or cisplatin plus 5-FU. However, the retrospective design of the current study must be taken into account when using these findings to recommend a chemotherapy regimen. Additional (matched-pair) studies should be performed directly comparing two radiochemotherapy programs.
In contrast to other significant predictors, the total radiation dose (70 Gy vs. 60-66 Gy) had no significant impact on LRC or OS. This finding may be due to the relatively small number of patients (n=38) who received doses of 60-66 Gy. However, it may also be true that a total dose of 66 Gy is sufficient for definitive radio(chemo)therapy of advanced SCCHN when administered with chemotherapy. A randomized trial comparing 66 Gy to 70 Gy for definitive treatment of SCCHN is recommended.
In conclusion, this study demonstrated the prognostic significance of the pre-radiotherapy hemoglobin levels for definitive treatment of SCCHN. Furthermore, concurrent chemotherapy is mandatory to achieve optimal results. Cisplatin alone with three courses of 100 mg/m2 (on days 1, 22 and 43) or two courses of 20 mg/m2 (on days 1-5 and 29-33) appear most effective. However, further studies are required to identify the optimal chemotherapy program. A total radiation dose of 70 Gy may not be superior to lower doses, which needs to be properly investigated in a future randomized trial. These findings, regarding prognostic factors, can aid treatment recommendations and stratification of groups when designing prospective trials.
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 March 10, 2016.
- Revision received April 11, 2016.
- Accepted April 13, 2016.
- Copyright© 2016 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved