Abstract
Background/Aim: Patients with unresectable head-and-neck cancer (SCCHN) unable to tolerate radiochemotherapy may receive unconventionally fractionated radiotherapy. This retrospective study compared both treatments. Patients and Methods: Eight patients unsuitable for chemotherapy were assigned to accelerated fractionation with concomitant boost (AF-CB, 69.6 Gy/39 fractions) over 5.5 weeks (group A) and 72 patients to cisplatin-based radiochemotherapy (70 Gy/35 fractions) over 7 weeks (group B). Groups were matched (cancer site, gender, age, performance score, T-/N-stage, histologic grade) and compared for loco-regional control (LRC), metastases-free survival (MFS), overall survival (OS) and toxicities. Results: LRC, MFS, OS and radiation-related toxicities were not significantly different between groups A and B. Improved outcomes were associated with favorable cancer site, better performance score and T3-stage. In group B, toxicity led to reduction/discontinuation of chemotherapy in 38.9% and interruptions of radiotherapy >7 days in 19.3% of patients. Conclusion: AF-CB appeared a reasonable alternative for patients who cannot safely receive radio-chemotherapy for unresectable SCCHN.
- Locally advanced head-and-neck cancer
- stage IV disease
- accelerated fractionation
- concomitant boost
- conventional radio-chemotherapy
Head-and-neck cancers represented the 7th most common malignancy worldwide in 2018 (1, 2). The vast majority of these cancers were squamous cell carcinomas (SCCHN). Locally advanced tumors may be unresectable and require definitive radiotherapy. The outcomes after definitive radiotherapy were significantly improved with the addition of concurrent (mainly platin-based) chemotherapy. In 2000, a randomized trial compared 66-72 Gy of radiotherapy alone to 66-72 Gy of radiotherapy with concurrent cisplatin and 5-fluorouracil (5-FU) for unresectable locally advanced SCCHN (3). The combined treatment resulted in significantly better loco-regional control (LRC). In 2004, another randomized trial compared 70 Gy in 35 fractions alone to the same regimen plus concurrent carboplatin/5-FU for definitive treatment of locally advanced oropharynx cancer (4). In this trial, concurrent radiochemotherapy was associated with significantly improved disease-free survival (DFS) and LRC and almost significantly improved overall survival (OS) when compared to radiotherapy alone. These results were confirmed in a large meta-analysis that included 93 randomized trials and demonstrated significantly better OS for concurrent radio-chemotherapy compared to radiation alone (5).
Unfortunately, the addition of chemotherapy to radiotherapy significantly increased grade ≥3 acute toxicities (6, 7). Thus, a considerable number of patients, particularly patients with significant co-morbidities and elderly patients, are unable to withstand concurrent radio-chemotherapy. Moreover, increased toxicity caused by the addition of chemotherapy may require interruptions of the radiotherapy, which can significantly impair the patients’ prognoses (8). Patients who are unsuitable for chemotherapy or other systemic agents are treated with radiotherapy alone. Treatment outcomes can be improved with unconventionally fractionated radiotherapy over less treatment time. According to a meta-analysis, unconventional fractionation was significantly superior to conventional fractionation (daily fractions of 2.0 Gy on five consecutive days per week) with respect to LRC and OS (9).
The term “unconventional fractionation” summarizes several altered dose-fractionation regimens including accelerated fractionation with concomitant boost (AF-CB) (9, 10). According to a randomized trial, AF-CB (72 Gy in 42 fractions over 6 weeks) resulted in more favorable outcomes than conventional fractionation (10). However, in this trial, AF-CB was associated with significantly increased late toxicity. Since the publication of the trial, other AF-CB programs were developed (11-14). One of these programs consisted of 69.9 Gy in 39 fractions given over 5.5 weeks (11). To our knowledge, this AF-CB program has not yet been compared to conventionally fractionated radio-chemotherapy. Therefore, the current study was initiated, which compared AF-CB to conventionally fractionated (70 Gy in 35 fractions) cisplatin-based radio-chemotherapy for definitive treatment of unresectable locally advanced SCCHN.
Patients and Methods
Eighty patients with histologically proven locally advanced unresectable SCCHN were included in this retrospective study, which was approved by the Ethics Committee at the University of Lübeck (20-454). Patients had non-metastatic stage IV disease; TNM-stages included T3 N2 M0 (n=25), T4 N0 M0 (n=8), T4 N1 M0 (n=12) and T4 N2 M0 (n=35) (15, 16). Patients with T1- or T2-tumors were not included to increase the homogeneity of the study population. The 7th edition of the American Joint Committee on Cancer staging manual was applied, since the status of human papilloma virus was not available for most patients, but is necessary for the staging of oropharynx cancers in the 8th edition (17).
Eight patients unable to receive radio-chemotherapy due to comorbidity were treated with accelerated radiotherapy including concomitant boost between 2011 and 2019 (group A). Initially, 30 Gy (15 daily fractions of 2.0 Gy over 3 weeks) were given to the primary tumor and regional lymph nodes including low-risk areas. After 30 Gy, the same volumes received additional 21.6 Gy with 1.8 Gy per fraction in the morning for 2.5 weeks (=12 treatment days). After an interval of ≥6 hours, which allowed the normal tissue to recover, 1.5 Gy were administered on the same days. The dose of 1.5 Gy was given to the primary tumor and high/intermediate-risk lymph node areas for 6 days (first concomitant boost, cumulative dose=60.6 Gy), and to the primary tumor and high-risk lymph node areas for another 6 days (second concomitant boost, cumulative dose=69.6 Gy). The treatment time including both concomitant boosts was 5.5 weeks (11, 18).
The other 72 patients (from an existing database) had been assigned to conventionally fractionated radiotherapy with concurrent cisplatin-based chemotherapy between 2000 and 2014 (group B). Initially, 50 Gy (daily fractions of 2.0 Gy over 5 weeks) were given to the primary tumor and regional lymph nodes including low-risk areas. Afterwards, a boost of 10 Gy (2.0 Gy per fraction, once daily) was administered to the primary tumor and high/intermediate-risk lymph node areas (cumulative dose=60 Gy), followed by a second boost of 10 Gy to the primary tumor and high-risk lymph node areas (cumulative dose=70 Gy). Thus, the treatment time in group B was 7 weeks. Concurrent cisplatin-based chemotherapy included either weekly administration of 30 mg/m2/d of cisplatin (n=15), 100 mg/m2 of cisplatin every 3 weeks (n=14), 20 mg/m2/d1-5 of cisplatin every 4 weeks (n=29), 20 mg/m2/d1-5 of cisplatin plus 600 mg/m2/d1-5 of 5-fluorouracil (5-FU) every 4 weeks (n=12) or 20 g/m2/d1-5 of cisplatin plus 1000 mg/m2/d1-5 of 5-FU every 4 weeks (n=2).
Both groups were matched for tumor site (oropharynx vs. hypopharynx vs. larynx), gender, age at radiotherapy (≤55 vs. >55 years), primary tumor stage (T3 vs. T4), stage of regional lymph nodes (N0-1 vs. N2) and histologic grade (G1-2 vs. G3). Matching for Karnofsky performance score (KPS 60-70 vs. 80-100) was not possible, since the KPS was significantly worse in the AF-CB group. The distributions of the parameters are shown in Table I (comparisons performed with Fisher’s exact test).
The groups were compared for LRC, metastases-free survival (MFS) and OS, calculated from the last day of radiotherapy. Univariate analyses for these endpoints were performed with the Kaplan–Meier method and the log-rank test. Characteristics found to be significant (p<0.05) or indicated a trend (p<0.10) were additionally evaluated for independence using a Cox proportional hazard model (multivariate analysis). Moreover, treatment groups A and B were compared for acute (oral mucositis, radiation dermatitis) and late (regional lymph edema, xerostomia) radiation-related toxicities using the Fisher’s exact test. Again, p-values <0.05 were considered significant and p-values <0.10 indicated a trend.
Results
Median follow-up periods were 18.5 months (range=0-70 months) in the entire cohort, 21.5 months (range=2-50 months) in group A and 18.5 months (range=0-70 months) in group B, respectively. On univariate analyses, improved LRC was significantly associated with favorable cancer sites (oropharynx or larynx, p=0.034), and a trend was found for KPS 80-100 (p=0.078) (Table II). The difference between group A and group B was not significant (p=0.85, Figure 1). In the subsequent multivariate analysis, KPS showed a trend [hazard ratio (HR)=1.97, 95% confidence interval (CI)=0.89-4.25, p=0.093)]; cancer site was not significant (HR=1.10, 95%CI=0.70-1.71, p=0.68).
Better MFS was significantly associated with KPS 80-100 (p=0.005) on univariate analyses and lower T-stage (T3) showed a trend (p=0.070) (Table III). MFS of treatment groups A and B was not significantly different (p=0.88, Figure 2). In the multivariate analysis of MFS, KPS was significant (HR=3.13, 95%CI=1.29-7.75, p=0.012); a trend was found for T-stage (HR=2.79, 95%CI=0.93-11.99, p=0.068).
Median OS-times were 44 months in the entire cohort, 42.5 months in group A and 50 months in group B, respectively. On univariate analyses (Table IV), improved OS was significantly associated with favorable cancer sites (oropharynx or larynx, p=0.015), and KPS 80-100 showed a trend (p=0.073). No significant association was found for the type of treatment (p=0.47, Figure 3). In the multivariate analysis of OS, KPS showed a trend (HR=1.83, 95%CI=0.91-3.58, p=0.089), cancer site was not significant (HR=1.01, 95%CI=0.67-1.49, p=0.96).
The comparisons of the treatment groups A and B for grade ≥2 and grade ≥3 radiation-related toxicities did not reveal any significant differences (Table V). In group A, one patient did not receive the planned total dose of 69.6 Gy due to acute toxicity. In group B, four patients received less than 70 Gy because of acute treatment-associated toxicity, and in three patients, radiotherapy was limited to 66 Gy due to other reasons.
Data regarding interruptions of radiotherapy >7 days were available for all patients of group A and 57 patients of group B, respectively. Interruptions >7 days were required in 0 patients (0%) of group A and 11 patients (19.3%) of group B, respectively (p=0.33). Chemotherapy could not be given as planned in 28 patients (38.9%) of group B. Toxicities leading to reduction or discontinuation of chemotherapy included impairment in renal function (n=12), severe oral mucositis (n=8), grade 3 nausea/vomiting (n=5), pneumonia (n=3) and other infections (n=2). Chemotherapy was reduced/discontinued in 8/15 patients (53.3%) receiving 30 mg/m2/d of cisplatin weekly, in 10/14 patients (71.4%) receiving 100 mg/m2 of cisplatin every 3 weeks, in 6/29 patients (30.7%) receiving 20 mg/m2/d1-5 of cisplatin every 4 weeks, in 3/12 patients (25.0%) receiving 20 mg/m2 of cisplatin plus 600 mg/m2 of 5-FU on days 1-5 every 4 weeks, and in 1/2 patients (50.0%) receiving 20 g/m2 of cisplatin plus 1000 mg/m2 of 5-FU on days 1-5 every 4 weeks, respectively.
Discussion
The prognoses of patients with locally advanced or metastatic SCCHN require improvement. A considerable number of studies were performed during recent years to contribute to this goal (19-24). Patients with locally advanced disease usually receive resection of the primary tumor and dissection of the regional lymph nodes followed by adjuvant radiotherapy or, in case of risk factors, radio-chemotherapy (6, 7, 25). Many patients cannot receive surgery, because the tumor is considered unresectable and/or they have significant co-morbidities. For unresectable SCCHN, cisplatin-based concurrent radio-chemotherapy with conventional fractionation (70 Gy in 35 fractions of 2.0 Gy over 7 weeks) is widely considered the standard treatment (3-5). However, due to co-morbidities including decreased renal function, peripheral neuropathy and hearing loss, a considerable number of patients cannot safely receive standard cisplatin-based chemotherapy. Other potential options of systemic therapy include carboplatin, mitomycin C plus 5-FU and epidermal growth factor receptor (EGFR) antibodies such as cetuximab (26-28). Since also these agents can be associated with significant side effects, the majority of patients unsuitable for cisplatin-based chemotherapy cannot receive other systemic treatments. Moreover, all systemic treatment increases the toxicity of the mandatory radiotherapy. This may lead to interruptions of the radiation treatment of more than 7 days, which has been shown to have a negative impact on the patients’ prognoses. In the multivariate analysis of a study in patients with non-metastatic stage IV SCCHN, lack of interruptions of radiotherapy >7 days was associated with improved LRC (risk ratio=3.32, p=0.015) and OS (risk ratio=2.59, p=0.021) (8).
For these patients, alternative treatment options are required that improve LRC and OS compared to conventionally fractionated radiotherapy alone. Improved outcomes were demonstrated for unconventional fractionated radiotherapy. In a randomized trial, hyper-fractionated radiotherapy (81.6 Gy in 68 fractions, i.e. 2×1.2 Gy per day on 5 days per week) and AF-CB (72 Gy in 42 fractions over 6 weeks, i.e. 30×1.8 Gy/day to a large field plus a concomitant boost of 1.5 Gy/day given 6 hours after the dose to the large field for the last 12 treatment days) resulted in significantly better LRC and DFS compared to conventional radiotherapy (70 Gy in 35 fractions) (10). The results of this trial were updated in 2014 (29). The absolute reduction in cumulative loco-regional failure at 5 years compared to conventional fractionation was 6.5% for hyper-fractionated radiotherapy and 6.6% for AF-CB, respectively. When considering the patients censored for loco-regional control at 5 years, p-values were 0.05 for hyper-fractionation and 0.11 for AF-CB, respectively. Both hyper-fractionation and AF-CB were significantly superior to conventional fractionation with respect to DFS (29). When considering the patients censored for DFS at 5 years, p-values were 0.01 for hyperfractionation and 0.05 for AF-CB, respectively. In addition, a meta-analysis of 15 trials demonstrated a benefit with respect to LRC for AF-CB (9). This meta-analysis also observed a survival benefit for unconventionally fractionated radiotherapy. The absolute benefit at 5 years was larger for hyper-fractionated radiotherapy than for AF-CB (8% vs. 2%). According to the authors of this meta-analysis, this difference should be interpreted with caution because of variation in patient characteristics between the treatment groups (9). A potential advantage of AF-CB compared to hyper-fractionation is the lower number of fractions (29-42 vs. 60-68 fractions). This can be particularly important for institutions with waiting lists due to high patient load or limited capacities at their linear accelerators.
After the trial of Fu et al. (10), additional AF-CB programs were reported that achieved promising results (11-14). The AF-CB program used in the current study was presented in a German trial in 2001 (11). The value of AF-CB has not yet been finally clarified. According to two meta-analyses, radiotherapy with accelerated fractionation such as AF-CB alone cannot entirely compensate for the lack of concurrent chemotherapy (30, 31). In a randomized phase III trial of 216 patients with oropharynx cancer, AF-CB (67.5 Gy in 40 fractions over 5 weeks) provided better compliance, toxicity profile and quality of life with similar disease control when compared to concurrent radio-chemotherapy including 66 Gy in 33 fractions over 6.5 weeks plus cisplatin 100mg/m2 on days 1, 22 and 43 (32). Moreover, in a recent randomized trial published in 2020, response rates and DFS were not significantly different for AF-CB and concurrent conventionally fractionated radio-chemotherapy (33). Thus, more studies comparing AF-CB and radio-chemotherapy for locally advanced SCCHN are warranted.
To our knowledge, the AF-CB included in the present study has not yet been compared to standard concurrent radio-chemotherapy with conventional fractionation. To allow better comparability of both treatments, the patients were matched for cancer site, gender, age, primary tumor stage, stage of regional lymph nodes and histologic grade. Matching for KPS was not possible, since KPS was significantly worse in group A. Patients receiving AF-CB were specifically unsuitable for chemotherapy due to significant co-morbidities that also impact KPS.
According to the results of this study, AF-CB was not significantly inferior to conventionally fractionated concurrent radio-chemotherapy with respect to LRC, MFS, OS and radiation-related toxicities. In contrast to the type of treatment, improved outcomes were associated with favorable cancer site (oropharynx or larynx), KPS of 80-100 and T3-stage (compared to T4-stage). These prognostic factors were also identified in previous studies demonstrating consistency of these findings to other studies (34-37). Limitations of this study include its retrospective design, the small sample size in group A, the lack of data regarding the status of the human papilloma virus, and the different treatment periods between the two treatment groups. In 15 patients (20.8%) of group B, chemotherapy consisted of weekly administration of 30 mg/m2/d of cisplatin. In two previous studies and a meta-analysis, weekly cisplatin appeared less effective than 100 mg/m2 of cisplatin every 3 weeks and 20 mg/m2/d1-5 of cisplatin every 4 weeks (38-40). Thus, the results after radio-chemotherapy might have been more favorable without patients receiving weekly cisplatin. Moreover, 38.9% of patients in group B did not receive their chemotherapy as planned due to acute toxicity. This demonstrates the importance of proper selection of patients for concurrent radio-chemotherapy. When patients with unresectable SCCHN undergo a careful selection process prior to treatment, the proportion of patients receiving the complete radio-chemotherapy as planned would likely increase. Concurrent radio-chemotherapy will remain the treatment-of-choice for the majority of patients with unresectable SCCHN.
In summary, given the limitations of this study, AF-CB produced promising results and is a reasonable alternative for the treatment of unresectable SCCHN in patients who cannot receive radio-chemotherapy. Confirmation of the results with a larger prospective trial including the AF-CB program used in the present study is warranted.
Footnotes
Authors’ Contributions
The study was designed by all Authors. The data of the new patients were collected by C.A.N. The analyses of the data used for this study were performed by S.E.S. and D.R. The article was drafted by D.R. and S.E.S. and approved by all Authors.
Conflicts of Interest
The Authors report no conflicts of interest related to the present study.
- Received November 26, 2020.
- Revision received December 3, 2020.
- Accepted December 4, 2020.
- Copyright© 2021, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.