Advertisement

The DAHANCA 6 randomized trial: Effect of 6 vs 5 weekly fractions of radiotherapy in patients with glottic squamous cell carcinoma

      Abstract

      Purpose

      The DAHANCA 6 trial evaluated tumor response and morbidity after moderate accelerated radiotherapy compared to conventional fractionated radiotherapy in patients treated for glottic squamous cell carcinoma (SCC). Further, the failure pattern and incidence of new primary tumors were explored.

      Patients and methods

      Six hundred and ninety-four patients with non-metastatic glottic SCC were randomized between six or five weekly fractions (fx/w) of radiotherapy to the same total dose. The median treatment time was 38 and 46 days, respectively. The primary endpoint was loco-regional failure.

      Results

      Median follow-up time was 14.5 years. Of the 177 failures, 167 involved T-site. The cumulative incidence of loco-regional failure (LRF) was 21.6% in the 6 fx/w group and 29.3% in the 5 fx/w group and the corresponding hazard rate (HR) of LRF was 0.72 (CI: 0.53–0.97, p = 0.04). The effect of acceleration on LRF was especially evident in well differentiated tumors (HR = 0.42 (CI: 0.23–0.75) and in T1–2 tumors (HR = 0.60 (CI: 0.41–0.89)). The HR of laryngectomy was 0.72 (CI: 0.50–1.04) in the 6 fx/w group compared to the 5 fx/w group. The hazards of disease-specific death, event-free survival, and overall survival were comparable between the two groups. Significantly more patients experienced severe acute mucositis in the 6 fx/w group but the incidence of late morbidity was comparable between the groups. New primary tumors occurred in 22.5% of the patients.

      Conclusion

      Moderate accelerated radiotherapy significantly improved loco-regional control in patients with glottic SCC.

      Keywords

      Glottic squamous cell carcinoma (SCC) is primarily a local disease and radiotherapy and/or surgery are used to achieve disease control. For decades the treatment of choice in Denmark has been radiotherapy. The efficacy of radiotherapy is related to different biological factors including intrinsic cellular radiosensitivity, tumor hypoxia, and stem cell proliferation during treatment. Accelerated tumor repopulation was described in head and neck SCC after initiation of radiotherapy [
      • Withers H.R.
      • Taylor J.M.
      • Maciejewski B.
      The hazard of accelerated tumor clonogen repopulation during radiotherapy.
      ]. Reduced overall treatment time is expected to counteract the accelerated growth and thereby improve loco-regional control [
      • Withers H.R.
      • Taylor J.M.
      • Maciejewski B.
      The hazard of accelerated tumor clonogen repopulation during radiotherapy.
      ,
      • Hansen O.
      • Overgaard J.
      • Hansen H.S.
      • Overgaard M.
      • Hoyer M.
      • Jorgensen K.E.
      • et al.
      Importance of overall treatment time for the outcome of radiotherapy of advanced head and neck carcinoma: dependency on tumor differentiation.
      ]. Such shorter overall treatment time without a dose reduction can be achieved either by applying a higher dose per fraction or by applying more fractions per week. In either situation, there is a risk of increasing the treatment related morbidity.
      In two parallel randomized trials, DAHANCA 6 and 7, the effect of moderate accelerated radiotherapy (ACC RT) on loco-regional control in patients with head and neck SCC was investigated [
      • Overgaard J.
      • Hansen H.S.
      • Specht L.
      • Overgaard M.
      • Grau C.
      • Andersen E.
      • et al.
      Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial.
      ]. The two studies were reported together. Due to prolonged inclusion in the DAHANCA 6 trial the median follow up in this cohort, was less than 3 years in the primary analysis. Subsequent sub-analyses revealed that ACC RT improved loco-regional control regardless of HPV/p16-status [
      • Lassen P.
      • Eriksen J.G.
      • Krogdahl A.
      • Therkildsen M.H.
      • Ulhoi B.P.
      • Overgaard M.
      • et al.
      The influence of HPV-associated p16-expression on accelerated fractionated radiotherapy in head and neck cancer: evaluation of the randomised DAHANCA 6&7 trial.
      ] and also improved loco-regional control in patients with high tumor expression of the epidermal growth factor receptor and well/moderate tumor differentiation [
      • Eriksen J.G.
      • Steiniche T.
      • Overgaard J.
      (DAHANCA) DH and NC study group. The influence of epidermal growth factor receptor and tumor differentiation on the response to accelerated radiotherapy of squamous cell carcinomas of the head and neck in the randomized DAHANCA 6 and 7 study.
      ]. A detailed analysis of treatment related morbidity demonstrated some increase in acute, but no increase in late morbidity in the ACC RT group [
      • Mortensen H.R.
      • Overgaard J.
      • Specht L.
      • Overgaard M.
      • Johansen J.
      • Evensen J.F.
      • et al.
      Prevalence and peak incidence of acute and late normal tissue morbidity in the DAHANCA 6&7 randomised trial with accelerated radiotherapy for head and neck cancer.
      ,
      • Mortensen H.R.
      • Overgaard J.
      • Jensen K.
      • Specht L.
      • Overgaard M.
      • Johansen J.
      • et al.
      Factors associated with acute and late dysphagia in the DAHANCA 6 & 7 randomized trial with accelerated radiotherapy for head and neck cancer.
      ].
      Though not all find an effect of acceleration [
      • Zackrisson B.
      • Nilsson P.
      • Kjellen E.
      • Johansson K.A.
      • Modig H.
      • Brun E.
      • et al.
      Two-year results from a Swedish study on conventional versus accelerated radiotherapy in head and neck squamous cell carcinoma–the ARTSCAN study.
      ,
      • Jackson S.M.
      • Weir L.M.
      • Hay J.H.
      • Tsang V.H.
      • Durham J.S.
      A randomised trial of accelerated versus conventional radiotherapy in head and neck cancer.
      ,
      • Olmi P.
      • Crispino S.
      • Fallai C.
      • Torri V.
      • Rossi F.
      • Bolner A.
      • et al.
      Locoregionally advanced carcinoma of the oropharynx: conventional radiotherapy vs. accelerated hyperfractionated radiotherapy vs. concomitant radiotherapy and chemotherapy–a multicenter randomized trial.
      ,

      Zackrisson B, Kjellén Elisabeth, Söderström K, Brun E, Nyman J NP. Five-year results from a Swedish study on conventional versus accelerated radiotherapy in head and neck squamous cell carcinoma – the ARTSCAN trial. Radiother Oncol 2015; Accepted for publication.

      ], other studies have substantiated the positive effect of moderate ACC RT in head and neck SCC [
      • Hliniak A.
      • Gwiazdowska B.
      • Szutkowski Z.
      • Kraszewska E.
      • Kukolowicz P.
      • Jarzabski A.
      • et al.
      A multicentre randomized/controlled trial of a conventional versus modestly accelerated radiotherapy in the laryngeal cancer: influence of a 1 week shortening overall time.
      ,
      • Horiot J.C.
      • Bontemps P.
      • van den Bogaert W.
      • Le Fur R.
      • van den Weijngaert D.
      • Bolla M.
      • et al.
      Accelerated fractionation (AF) compared to conventional fractionation (CF) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomized trial.
      ], but have often not included early stage and/-or glottic tumors [
      • Beitler J.J.
      • Zhang Q.
      • Fu K.K.
      • Trotti A.
      • Spencer S.A.
      • Jones C.U.
      • et al.
      Final results of local-regional control and late toxicity of RTOG 9003: a randomized trial of altered fractionation radiation for locally advanced head and neck cancer.
      ,
      • Skladowski K.
      • Maciejewski B.
      • Golen M.
      • Pilecki B.
      • Przeorek W.
      • Tarnawski R.
      Randomized clinical trial on 7-day-continuous accelerated irradiation (CAIR) of head and neck cancer – report on 3-year tumour control and normal tissue toxicity.
      ]. The DAHANCA 6 study is, by far, the largest study specifically investigating the effect of moderate acceleration in glottic SCC patients. Most patients in the DAHANCA 6 study had early stage SCC and by now, the follow-up time is considerable.
      The aim of the present report was to describe the long-term outcome after six compared to five weekly fractions of radiotherapy in patients with glottic SCC. Further, the pattern of failure and the incidence of new primary tumors (NPT) were evaluated.

      Materials and methods

      The DAHANCA 6 study was described in detail by Overgaard et al. [
      • Overgaard J.
      • Hansen H.S.
      • Specht L.
      • Overgaard M.
      • Grau C.
      • Andersen E.
      • et al.
      Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial.
      ]. In brief, the DAHANCA 6 multicenter trial randomized patients to either six (6 fx/w) or five (5 fx/w) weekly fractions of radiotherapy. Eligibility criteria were stage I–IV (TNM-classification, UICC 2nd edition, 1987) biopsy proven SCC of the glottic larynx with no distant metastases and no previous treatment for the disease. Randomization was stratified according to gender, T classification (T1–2 or T3–4) and treatment center.
      A total of 694 patients were included for randomization between January 1992 and December 1999 (Fig. 1), of whom 690 were eligible for analysis. The distribution of tumor and patient characteristics was similar in the two groups (Table 1).
      Figure thumbnail gr1
      Fig. 1Trial overview. Percentages refer to the total number in the six and five weekly fraction groups, respectively. Abbreviations: 6 fx/w and 5 fx/w, 5 and 6 weekly fractions of radiotherapy, respectively; H&N, head and neck. *Includes suicide, accidents and treatment complications.
      Table 1Patient and tumor characteristics.
      Because of rounding not all percentages total to 100.
      6 fx/w5 fx/wTotalChi2-test
      n%n%p
      Included
      Patient nationality was 639 Danish, 46 Norwegian, 8 Swedish.
      349(100)341(100)690
      Age
       Median age (years)64650.55
      Wilcoxon rank sum test.
       Range (years)25–8734–85
       <5571(20)72(21)143
       55–65117(34)99(29)216
       >65161(46)170(50)3310.44
      Sex
       Male320(92)317(93)637
       Female29(8)24(7)530.53
      Performance Status
       0259(74)243(71)502
       183(24)77(23)160
       27(2)21(6)280.02
      TNM stage (UICC87)
       I196(56)181(53)377
       II104(30)103(30)207
       III44(13)54(16)98
       IV5(1)3(1)80.58
      T classification
       T1196(56)181(53)377
       T2105(30)110(32)215
       T342(12)49(14)91
       T46(2)1(0)70.19
      N classification
       0346(99)330(97)676
       >03(1)11(3)140.07
      Differentiation
       Well110(32)119(35)229
       Moderate153(44)143(42)296
       Poor52(14)38(11)90
       Unknown34(10)41(12)750.39
      T1N0 well differentiated
      67(19)65(19)1320.96
      Abbreviations: 6 fx/w and 5 fx/w: 6 and 5 weekly fractions of radiotherapy, respectively; n = absolute number.
      a Patient nationality was 639 Danish, 46 Norwegian, 8 Swedish.
      b Because of rounding not all percentages total to 100.
      c Wilcoxon rank sum test.
      Radiotherapy was delivered as opposing lateral fields using 4–6 MV photons (Suppl. Document 1 and 2). For node negative T1–4 tumors the target area was limited by the upper border of the thyroid cartilage and the caudal border of the cricoid cartilage with no elective lymph node irradiation. In N+ disease the more caudal node station was included for elective irradiation. Fractions of 2 Gy were given to a total of 68 Gy if the tumor and/or lymphnode diameter was larger than 4 cm, 66 Gy if the largest diameter was ⩽ to 4 cm, and 62 Gy in case of a T1 N0 well differentiated tumor.
      When assigned 6 fractions a week, by far, most of the patients received the 6th dose on a working day with at least 6 h separating the two daily fractions. In a few cases, the extra dose was provided on Saturdays. Twelve patients, 6 in each arm, did not receive the planned total dose. The median treatment time was 38 (range 32–74) and 46 days (range 37–71) in the 6- and 5 fx/w group, respectively.

      Follow-up

      The study population was followed routinely for at least 5 years. To ascertain long-term follow up, the survival status of all patients was updated by July 2013 using the Danish Central Population Registry. The cause of death registered in the DAHANCA-database was matched with information from the Danish Registry of Cause of Death. Cases of unknown failure status and death from or with glottic SCC were confirmed with an extensive national based chart review. The National Cancer Registry was used to obtain information on new primary tumors (NPT). Since a full update of NPT information was not available for non-Danish patients, these patients were excluded from the analysis of NPT incidence.

      Statistics and endpoint definitions

      The primary endpoint was loco-regional failure (LRF) defined as persistent or recurrent disease in tumor (T) and/or lymphnode (N) sites. All presence of SCC in the primary site was considered a recurrence regardless of the time span since the treatment unless otherwise stated. Secondary endpoints included: local failure (LF) defined as persistent or recurrent disease in T site; disease-specific death (DSD) defined as death from or with glottic cancer; overall survival (OS) defined as no death from any cause; event-free survival (EFS) defined as no persistent or recurrent disease or death from any cause; acute and late morbidity; and the incidence of salvage laryngectomy.
      All patients were followed from the day of randomization until death or July 1st 2013, except 8 patients who emigrated. All analyses were by intention to treat. Patient and tumor characteristics and acute morbidity were compared using χ2 or Wilcoxon rank sum test for binary and continuous parameters, respectively. Difference in late morbidity according to treatment was described using relative risk (RR). The Cox proportional hazards model was used to assess hazard ratios (HR) and presented as 6 compared to 5 fx/w when comparing treatments. Corresponding p-values were calculated with the Wald test. In the multivariate analysis of LRF, a test for interaction was performed to compare the effect of acceleration dependent on tumor differentiation and disease stages. All tests were two-sided and performed at a 5% significance level with 95% confidence intervals (CI).
      Cumulative incidence (CIP) estimates were adjusted for competing risks. Competing risks were: LRF (distant metastasis or death); LF (failures not including T-site or death); DSD (death in patients cured for the glottic cancer); Laryngectomy (death); NPT (death); OS and EFS (no competing events). Risk differences (RD) were estimated using the pseudo value approach [

      Zackrisson B, Kjellén Elisabeth, Söderström K, Brun E, Nyman J NP. Five-year results from a Swedish study on conventional versus accelerated radiotherapy in head and neck squamous cell carcinoma – the ARTSCAN trial. Radiother Oncol 2015; Accepted for publication.

      ]. RDs were presented as the percentage point lower risk after 6 fx/w compared to 5 fx/w. HR and RD estimates were made for 10-years of follow-up if nothing else was stated.

      Results

      The median follow-up regarding LRF was 14.5 years (range 4.3–21.5) for patients still alive, and 4.6 years (range 0.04–21.2) for patients who died.
      In total 177 patients experienced a failure, of which 13 patients never achieved primary control (Fig. 1). The T-site was involved in 167 failures; 150 T-site alone, 15 TN, one TNM, and one TM failure, respectively. The pattern of failure is described in detail in the Suppl. Fig. 1. Sixteen failures occurred later than 5 years after randomization, hereof 3 later than 10 years. The late recurrences were isolated T-site failures, except from one TN-failure.

      Locoregional failure

      Overall, the cumulative incidence of LRF was 25.4% (22.2–28.7) (Suppl. Table 1). On univariate analysis of the stratification parameters a significantly higher hazard of LRF in T3–4 compared to T1–2 tumors (HR = 6.2 (CI: 4.6–8.6)) was observed.
      The effect of acceleration was expressed as the difference in LRF between the two treatment groups. The cumulative incidence of LRF was 22% (CI: 17–26%) in the 6 fx/w group and 29% (CI: 25–34%) in the 5 fx/w group, respectively, with a corresponding RD of 7.8% (CI: 1.2–14.3) and a HR of 0.72 (CI: 0.53–0.97, p = 0.04) (Fig. 2).
      Figure thumbnail gr2
      Fig. 2Loco-regional failure, local failure, disease specific death and overall survival in the two treatment groups. Cumulative incidence curves (corrected for competing risks, see text for details) showing loco-regional failure (A), disease specific death (B), local failure in the whole group (D), local failure in T1–2 tumors (E), and local failure in T3–4 tumors (F), respectively. (C) Shows overall survival presented as a Kaplan Meier estimate. Abbreviations: 6 fx/w and 5 fx/w, 6 and 5 weekly fractions of radiotherapy, respectively; RD, Risk difference estimates showing the absolute lower risk of event in the 6 fx/w group compared to the 5 fx/w group; HR, Hazard rate was calculated as the relative hazard of event in the 6 fx/w group compared to the 5 fx/w group; CI, 95% confidence interval.
      A subgroup analysis showed a distinct effect of acceleration on LRF in well differentiated tumors (HR = 0.42 (CI: 0.23–0.75)) and T1–2 tumors (HR = 0.6 (CI: 0.41–0.89)) (Fig. 3, Suppl. Tables 1 and 2).
      Figure thumbnail gr3
      Fig. 3Subgroup analysis of loco-regional failure. Event: Loco-regional failure (LRF) was defined as persistent or recurrent disease in the tumor and/or lymphnode sites (absolute number in subgroup displayed). Total: The number of patients in a specific subgroup. All estimates are 10-year values. Abbreviations: 6 fx/w and 5 fx/w, 6 and 5 weekly fractions of radiotherapy, respectively; HR, Hazard rate was calculated as the relative hazard of LRF in the 6 fx/w group compared to the 5 fx/w group; CIP, Cumulative incidence estimates (corrected for competing risks being metastatic recurrence and death); RD, Risk difference estimates showing the absolute lower risk of LRF in the 6 fx/w group compared to the 5 fx/w group; 95% CI, 95% confidence intervals.
      In a multivariate analysis the adjusted HR of LRF was 0.77 (CI: 0.57–1.04) in the 6 fx/w group compared to the 5 fx/w group. (Suppl. Table 3). The covariates included tumor stage I–II vs III–IV, differentiation (well, moderate, poor, unknown), age (<55, 55–65, >65), and performance status (0 or >0). A test for interaction comparing the effect of acceleration in the four tumor differentiation groups demonstrated a significant difference in the response to acceleration between the differentiation groups. Thus, the multivariate analysis was performed for each differentiation group separately. ACC RT reduced the incidence of LRF more with more mature differentiation (Suppl. Table 3).

      Local failure

      Considering the predominance of local T-site failure, the treatment effect was analyzed using this endpoint (Fig. 2D). The incidence of LF was 8% (CI: 2–15%) lower in the 6 fx/w group compared to the 5 fx/w group, and the corresponding HR was 0.70 (CI: 0.51–0.95)). The effect of acceleration was observed in the group of patients with T1–2 tumors (Fig. 2E) with a HR of 0.60 (CI: 0.40–0.89) compared with a HR of 1.0 (CI: (−16.1)–22.5) in T3–4 tumors (Fig. 2F). Due to the low number of failures involving N-site, a meaningful analysis could not be performed for N-site failure (Suppl. Table 4).

      Survival

      At the time of analysis 77% (n = 529) of patients were dead (Fig. 1). The main cause of death was cancer (n = 257). Glottic cancer caused 102 deaths. Other diseases caused 221 deaths. Forty-five patients died from unknown reasons and eight patients were lost to follow up due to emigration.
      The cumulative incidence of disease specific death (DSD) was significantly lower in the 6 fx/w group (12%) than in the 5 fx/w group (17%), but the HR 0.69 (CI: 0.46–1.02, p = 0.07) did not reach statistical significance (Fig. 2B). There was no difference in event-free survival (HR = 0.91 (CI: 0.76–1.08)) (Suppl. Fig. 2) and overall survival (HR = 0.98 (CI: 0.82–1.17)) (Fig. 2C).

      Morbidity

      The acute morbidity was more pronounced in the 6 fx/w group, e.g. a significant increase in confluent mucositis was found in the 6 fx/w group, particularly in week 3–6 (p < 0.05) (Fig. 4). Still 98% of patients in both treatment groups completed the treatment and all acute morbidity resolved within three months after the start of radiotherapy. No difference in the frequency of moderate to severe late morbidity was observed between the groups as described by Mortensen [
      • Mortensen H.R.
      • Overgaard J.
      • Specht L.
      • Overgaard M.
      • Johansen J.
      • Evensen J.F.
      • et al.
      Prevalence and peak incidence of acute and late normal tissue morbidity in the DAHANCA 6&7 randomised trial with accelerated radiotherapy for head and neck cancer.
      ], except from loss of taste which was more frequent in the 5 fx/w group (Fig. 4, Suppl. Fig. 4 and Table 5).
      Figure thumbnail gr4
      Fig. 4Morbidity. Left: acute morbidity. Right: relative risk of late (>6 month after treatment) morbidity. Event = number of patients with moderate to severe morbidity. All = total number of observations. For further description of morbidity parameters see and
      [
      • Mortensen H.R.
      • Overgaard J.
      • Specht L.
      • Overgaard M.
      • Johansen J.
      • Evensen J.F.
      • et al.
      Prevalence and peak incidence of acute and late normal tissue morbidity in the DAHANCA 6&7 randomised trial with accelerated radiotherapy for head and neck cancer.
      ]
      .

      Salvage treatment

      One hundred and twenty-eight patients with failure were considered suitable for curatively intended salvage surgery leading to persistent disease control in 75 patients; 72 failures included T-site, three were solitary lymphnode or metastatic recurrences (Suppl. Fig. 1). Salvage was successful in 48% of patients with T1–2 tumor (55 of 114 failures) and 31% of patients with T3–4 tumor (20 of 63 failures) (Fig. 1). Salvage laryngectomy was used for 51 and 67 patients in the 6- and 5 fx/w groups, respectively. The cumulative incidence of laryngectomy was 14% (11–18%) and 19% (15–24%) in the 6- and 5 fx/w groups, respectively, with a HR of 0.72 (CI: 0.50–1.04). The median survival time after successful salvage was 8.4 (5.3–11.7) years (Suppl. Fig. 3).

      New primary tumors

      Information on new primary tumors was available for 639 patients. One hundred and eighty-three patients developed one NPT, and 16 patients developed two (Fig. 5). Non-malignant melanoma skin cancer and cancer diagnosed prior to the glottic SCC were not included. After 10 years of observation the cumulative incidence of the first NPT was 22.6% (CI: 19.4–25.9%). New primary HNSCC was seen in 1.1%, 2.4%, and 3.4% of patients after 5, 10, and 15 years of follow-up, respectively. There was no difference in the incidence of NPTs between the two treatment groups. The 10 year CIP of NPT was 21.5% (CI: 17.2–26.2) in the 6 fx/w group and 23.7% (CI: 19.1–28.5) in the 5 fx/w group, respectively, with a corresponding RD of 2.1% (CI: (−4.4)–8.6) and a HR of 0.89 (CI: 0.67–1.17).
      Figure thumbnail gr5
      Fig. 5New primary tumor incidence. Since a full update of NPT information was not available for non-Danish patients, only Danish patients (639) were included in the analysis. New primary tumor sites included: lung and trachea 72, gastro intestinal 40, kidney and bladder 25, head and neck 23, prostate 20, hematological 13, breast 3, pancreas 5, malignant melanoma 2, brain 1, liver 1, and unknown primary site 10. The 10-year cumulative incidence of lung- and new primary head and neck cancer is shown below the curve.

      Discussion

      In the largest randomized radiotherapy study concerning only glottic SCC, long term follow-up confirms that moderately accelerated radiotherapy significantly reduced the risk of loco-regional and local failure compared to conventional fractionation. The MARCH meta-analysis, which included the early report of the DAHANCA 6 study, investigated the effect of different accelerated fractionated regimens in head and neck SCC [
      • Bourhis J.
      • Overgaard J.
      • Audry H.
      • Ang K.K.
      • Saunders M.
      • Bernier J.
      • et al.
      Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta-analysis.
      ]. A five year reduction of 18% and 8% in the hazard of loco-regional failure and death was reported, respectively. Apart from the DAHANCA 6 trial, most studies included in the MARCH analysis reported outcome on mixed cancer site cohorts and only few included early glottic SCC [
      • Hliniak A.
      • Gwiazdowska B.
      • Szutkowski Z.
      • Kraszewska E.
      • Kukolowicz P.
      • Jarzabski A.
      • et al.
      A multicentre randomized/controlled trial of a conventional versus modestly accelerated radiotherapy in the laryngeal cancer: influence of a 1 week shortening overall time.
      ,
      • Horiot J.C.
      • Bontemps P.
      • van den Bogaert W.
      • Le Fur R.
      • van den Weijngaert D.
      • Bolla M.
      • et al.
      Accelerated fractionation (AF) compared to conventional fractionation (CF) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomized trial.
      ].
      In the DAHANCA 6 study, the efficacy of acceleration was most pronounced in early (T1–2) tumors, representing 86% of the included carcinomas. The 10-year risk difference in loco-regional and local failure in the T1–2 group was 9% and 8%, respectively. The Polish KBN PO 79 trial also studied the effect of moderately ACC RT in 395 T1–3 glottic SCC. The 5% (24% vs 19%, p > 0.05) difference in absolute risk of loco-regional failure observed after 3 years was in line with our findings [
      • Hliniak A.
      • Gwiazdowska B.
      • Szutkowski Z.
      • Kraszewska E.
      • Kukolowicz P.
      • Jarzabski A.
      • et al.
      A multicentre randomized/controlled trial of a conventional versus modestly accelerated radiotherapy in the laryngeal cancer: influence of a 1 week shortening overall time.
      ]. In a Dutch retrospective analysis of 1050 T1-2N0 glottic SCC, treatment with 6 fx/w was associated with a 6% reduction in local failure after 5 years compared to 5 fx/w [
      • Al-Mamgani A.
      • van Rooij P.H.
      • Woutersen D.P.
      • Mehilal R.
      • Tans L.
      • Monserez D.
      • et al.
      Radiotherapy for T1–2N0 glottic cancer: a multivariate analysis of predictive factors for the long-term outcome in 1050 patients and a prospective assessment of quality of life and voice handicap index in a subset of 233 patients.
      ]. Accelerated hypofractionated radiotherapy reduced the 5 years risk of LF with 15% (23% vs 8%, p < 0.05) compared to conventional fractionated radiotherapy in a Japanese study of 180 T1 patients with glottic SCC. Hypofractionation was associated with a 3 Gy reduction in total treatment dose [
      • Yamazaki H.
      • Nishiyama K.
      • Tanaka E.
      • Koizumi M.
      • Chatani M.
      Radiotherapy for early glottic carcinoma (T1N0M0): results of prospective randomized study of radiation fraction size and overall treatment time.
      ]. A similar Korean trial of 156 patients with T1–T2 glottic scc substantiated the finding of improved disease control after accelerated hypofractionated radiotherapy compared to conventional fractionated radiotherapy [
      • Moon S.H.
      • Cho K.H.
      • Chung E.J.
      • Lee C.G.
      • Lee K.C.
      • Chai G.Y.
      • et al.
      A prospective randomized trial comparing hypofractionation with conventional fractionation radiotherapy for T1–2 glottic squamous cell carcinomas: results of a Korean Radiation Oncology Group (KROG-0201) study.
      ]. In the RTOG 9512 study hyperfractionation with a 9.2 Gy dose increase and 0.5 week treatment time reduction increased local control 8% (78% vs 70%, p > 0.05) in T2N0 glottic cancer patients compared to conventional RT [
      • Trotti A.
      • Zhang Q.
      • Bentzen S.M.
      • Emami B.
      • Hammond M.E.
      • Jones C.U.
      • et al.
      Randomized trial of hyperfractionation versus conventional fractionation in T2 squamous cell carcinoma of the vocal cord (RTOG 9512).
      ]. Overall, the evidence of a positive effect of reduced treatment time on loco-regional control is convincing.
      The benefit of accelerated fractionation was most pronounced in well and moderate differentiated tumors. This finding is in agreement with previous studies [
      • Hansen O.
      • Overgaard J.
      • Hansen H.S.
      • Overgaard M.
      • Hoyer M.
      • Jorgensen K.E.
      • et al.
      Importance of overall treatment time for the outcome of radiotherapy of advanced head and neck carcinoma: dependency on tumor differentiation.
      ]. This indicates that accelerated repopulation demands a functional cellular behavior of the tumor cells in order to respond adequately to the induced trauma, and such mechanisms may require a certain degree of cellular differentiation [
      • Hansen O.
      • Overgaard J.
      • Hansen H.S.
      • Overgaard M.
      • Hoyer M.
      • Jorgensen K.E.
      • et al.
      Importance of overall treatment time for the outcome of radiotherapy of advanced head and neck carcinoma: dependency on tumor differentiation.
      ,
      • Overgaard J.
      • Alsner J.
      • Eriksen J.
      • Horsman M.R.
      • Grau C.
      Importance of overall treatment time for the response to radiotherapy in patients with squamous cell carcinoma of the head and neck.
      ,
      • Dische S.
      • Saunders M.
      • Barrett A.
      • Harvey A.
      • Gibson D.
      • Parmar M.
      A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer.
      ].
      A positive dose–response relation in RT is well described for laryngeal cancer [
      • Overgaard J.
      • Hjelm-Hansen M.
      • Johansen L.V.
      • Andersen A.P.
      Comparison of conventional and split-course radiotherapy as primary treatment in carcinoma of the larynx.
      ]. That we only treated well differentiated T1 tumors with 62 Gy may therefore not expose the full benefit of accelerated fractionation to this tumor group. In absolute numbers, the possible effect of a dose increase is limited, though, since only 19 failures occurred in 139 patients with T1 well differentiated tumors (Suppl. Table 2).
      Almost two thirds of patients with a T3–4 tumor suffered from failure regardless of the treatment regimen indicating a need of intensified treatment. This may be achieved by escalating the dose by hyperfractionation [
      • Bourhis J.
      • Overgaard J.
      • Audry H.
      • Ang K.K.
      • Saunders M.
      • Bernier J.
      • et al.
      Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta-analysis.
      ] (to 76 Gy as recommended in the current DAHANCA guidelines), or by chemoradiotherapy [
      • Blanchard P.
      • Baujat B.
      • Holostenco V.
      • Bourredjem A.
      • Baey C.
      • Bourhis J.
      • et al.
      Meta-analysis of chemotherapy in head and neck cancer (MACH-NC): a comprehensive analysis by tumour site.
      ,
      • Bentzen J.
      • Toustrup K.
      • Eriksen J.G.
      • Primdahl H.
      • Andersen L.J.
      • Overgaard J.
      Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study.
      ], both regimens including the addition of hypoxic modification [
      • Overgaard J.
      • Hansen H.S.
      • Overgaard M.
      • Bastholt L.
      • Berthelsen A.
      • Specht L.
      • et al.
      A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) Protocol 5–85.
      ], which may add further benefit to large glottic tumors [
      • Bentzen J.
      • Toustrup K.
      • Eriksen J.G.
      • Primdahl H.
      • Andersen L.J.
      • Overgaard J.
      Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study.
      ]. Hypoxic modification was not included in the DAHANCA 6 study because a previously DAHANCA study showed no benefit of hypoxic modification in small glottic tumors [
      • Overgaard J.
      • Hansen H.S.
      • Andersen A.P.
      • Hjelm-Hansen M.
      • Jorgensen K.
      • Sandberg E.
      • et al.
      Misonidazole combined with split-course radiotherapy in the treatment of invasive carcinoma of larynx and pharynx: report from the DAHANCA 2 study.
      ].
      Primary surgery may represent an alternative treatment strategy for advanced glottic tumors. No survival benefit and a lower voice preservation were found when comparing primary surgery to conservative treatment in advanced laryngeal SCC [
      TD of VALCS Group
      Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer.
      ,
      • Groome P.A.
      • O’Sullivan B.
      • Irish J.C.
      • Rothwell D.M.
      • Math K.S.
      • Bissett R.J.
      • et al.
      Glottic cancer in Ontario, Canada and the SEER areas of the United States. Do different management philosophies produce different outcome profiles?.
      ].
      Accelerated RT did not influence survival in the DAHANCA 6 study. The absolute number of loco-regional failures in DAHANCA 6 was relatively low and 40% of the failures were successfully salvaged. Besides death from the glottic cancer, this patient group was prone to die from other diseases. Bøje et al. described that 38% of 3949 patients with laryngeal cancer suffered from important comorbidity [
      • Boje C.R.
      • Dalton S.O.
      • Gronborg T.K.
      • Primdahl H.
      • Kristensen C.A.
      • Andersen E.
      • et al.
      The impact of comorbidity on outcome in 12 623 Danish head and neck cancer patients: a population based study from the DAHANCA database.
      ]. Moreover, in the present study 23% of patients were diagnosed with a new primary tumor. ACC RT decreased the absolute risk of laryngectomy by 5% offering patients an improved quality of life compared to conventional fractionation.
      Glottic SCC is a local disease [
      • Jakobsen J.
      • Hansen O.
      • Jorgensen K.E.
      • Bastholt L.
      Lymph node metastases from laryngeal and pharyngeal carcinomas–calculation of burden of metastasis and its impact on prognosis.
      ]. The frequency of node positive disease at diagnosis was described in large cohort studies to be within 1% and 3% [
      • Jakobsen J.
      • Hansen O.
      • Jorgensen K.E.
      • Bastholt L.
      Lymph node metastases from laryngeal and pharyngeal carcinomas–calculation of burden of metastasis and its impact on prognosis.
      ,
      • Johansen L.V.
      • Grau C.
      • Overgaard J.
      Glottic carcinoma–patterns of failure and salvage treatment after curative radiotherapy in 861 consecutive patients.
      ] similar to the 2% in the DAHANCA 6 trial. This long-term follow up confirms that glottic SCC continues to be a local disease after primary radiotherapy treatment with only few regional (14%) and almost no (2%) distant failures. In the RTOG 9003 study concerning advanced HNSCC, only 3% of all failures occurred after 5 years of follow-up [
      • Beitler J.J.
      • Zhang Q.
      • Fu K.K.
      • Trotti A.
      • Spencer S.A.
      • Jones C.U.
      • et al.
      Final results of local-regional control and late toxicity of RTOG 9003: a randomized trial of altered fractionation radiation for locally advanced head and neck cancer.
      ]. In the DAHANCA 6 trial 10% of failures occurred after 5 years of follow-up. All recurrences occurring after 5 years were in patients with initial T1–2 tumors. This indicates a different recurrence profile in early and advanced HNSCC with a presentation of later recurrences or new primary SCC, potentially radiation induced, in early glottic SCC. Alternatively, this may reflect that more patients with early tumors become longtime survivors susceptible to late failures.
      Evaluation of the study outcome was limited by the absence of smoking data since smoking affects the efficacy of radiotherapy, and the risk of recurrence [
      • Al-Mamgani A.
      • van Rooij P.H.
      • Woutersen D.P.
      • Mehilal R.
      • Tans L.
      • Monserez D.
      • et al.
      Radiotherapy for T1–2N0 glottic cancer: a multivariate analysis of predictive factors for the long-term outcome in 1050 patients and a prospective assessment of quality of life and voice handicap index in a subset of 233 patients.
      ,
      • Hoff C.M.
      Importance of hemoglobin concentration and its modification for the outcome of head and neck cancer patients treated with radiotherapy.
      ]. The problem was modest, though, since the randomized study setup was likely to have allocated smokers evenly in the two treatment groups.
      The applied accelerated RT regimen has since proven feasible worldwide [
      • Overgaard J.
      • Mohanti B.K.
      • Begum N.
      • Ali R.
      • Agarwal J.P.
      • Kuddu M.
      • et al.
      Five versus six fractions of radiotherapy per week for squamous-cell carcinoma of the head and neck (IAEA-ACC study): a randomised, multicentre trial.
      ]. Further, the side effect profile of accelerated RT was highly acceptable with no difference in long-term side effects and only a moderate increase in acute side effects that did not reduce compliance to treatment [
      • Mortensen H.R.
      • Overgaard J.
      • Specht L.
      • Overgaard M.
      • Johansen J.
      • Evensen J.F.
      • et al.
      Prevalence and peak incidence of acute and late normal tissue morbidity in the DAHANCA 6&7 randomised trial with accelerated radiotherapy for head and neck cancer.
      ,
      • Overgaard J.
      • Mohanti B.K.
      • Begum N.
      • Ali R.
      • Agarwal J.P.
      • Kuddu M.
      • et al.
      Five versus six fractions of radiotherapy per week for squamous-cell carcinoma of the head and neck (IAEA-ACC study): a randomised, multicentre trial.
      ]. The effectiveness and feasibility of the DAHANCA schedule of 6 fractions per week has led to its adoption as the new standard fractionation in the RTOG 0522 and 1016 trials using IMRT [
      • Trotti A.
      • Machtay M.
      RTOG 9003: Legacies of a Landmark Trial.
      ,
      • Rosenthal D.I.
      • Fuller C.D.
      • Peters L.J.
      • Thames H.D.
      Final Report of Radiation Therapy Oncology Group Protocol 9003: provocative, but limited conclusions from exploratory analyses.
      ]. The use of IMRT or 3D conformal technique may further lower the late morbidity due to sparing of e.g. the contralateral arytenoid and esophagus compared to the opposing lateral fields used in the DAHANCA 6 trial. Alternatively, modern radiotherapy techniques hold the potential of dose escalation without an increase in morbidity.
      In conclusion, accelerated fractionation with six fractions per week should be preferred over conventional radiotherapy in the treatment of glottic cancer. In advanced disease, the treatment should be further intensified to reduce the high incidence of failure seen regardless of accelerated fractionation.

      Conflict of interest

      None.

      Acknowledgment

      The study was supported by the Danish Cancer Society .

      Appendix A. Supplementary data

      References

        • Withers H.R.
        • Taylor J.M.
        • Maciejewski B.
        The hazard of accelerated tumor clonogen repopulation during radiotherapy.
        Acta Oncol. 1988; 27: 131-146
        • Hansen O.
        • Overgaard J.
        • Hansen H.S.
        • Overgaard M.
        • Hoyer M.
        • Jorgensen K.E.
        • et al.
        Importance of overall treatment time for the outcome of radiotherapy of advanced head and neck carcinoma: dependency on tumor differentiation.
        Radiother Oncol. 1997; 43: 47-51
        • Overgaard J.
        • Hansen H.S.
        • Specht L.
        • Overgaard M.
        • Grau C.
        • Andersen E.
        • et al.
        Five compared with six fractions per week of conventional radiotherapy of squamous-cell carcinoma of head and neck: DAHANCA 6 and 7 randomised controlled trial.
        Lancet. 2003; 362: 933-940
        • Lassen P.
        • Eriksen J.G.
        • Krogdahl A.
        • Therkildsen M.H.
        • Ulhoi B.P.
        • Overgaard M.
        • et al.
        The influence of HPV-associated p16-expression on accelerated fractionated radiotherapy in head and neck cancer: evaluation of the randomised DAHANCA 6&7 trial.
        Radiother Oncol. 2011; 100: 49-55
        • Eriksen J.G.
        • Steiniche T.
        • Overgaard J.
        (DAHANCA) DH and NC study group. The influence of epidermal growth factor receptor and tumor differentiation on the response to accelerated radiotherapy of squamous cell carcinomas of the head and neck in the randomized DAHANCA 6 and 7 study.
        Radiother Oncol. 2005; 74: 93-100
        • Mortensen H.R.
        • Overgaard J.
        • Specht L.
        • Overgaard M.
        • Johansen J.
        • Evensen J.F.
        • et al.
        Prevalence and peak incidence of acute and late normal tissue morbidity in the DAHANCA 6&7 randomised trial with accelerated radiotherapy for head and neck cancer.
        Radiother Oncol. 2012; 103: 69-75
        • Mortensen H.R.
        • Overgaard J.
        • Jensen K.
        • Specht L.
        • Overgaard M.
        • Johansen J.
        • et al.
        Factors associated with acute and late dysphagia in the DAHANCA 6 & 7 randomized trial with accelerated radiotherapy for head and neck cancer.
        Acta Oncol. 2013; 52: 1535-1542
        • Zackrisson B.
        • Nilsson P.
        • Kjellen E.
        • Johansson K.A.
        • Modig H.
        • Brun E.
        • et al.
        Two-year results from a Swedish study on conventional versus accelerated radiotherapy in head and neck squamous cell carcinoma–the ARTSCAN study.
        Radiother Oncol. 2011; 100: 41-48
        • Jackson S.M.
        • Weir L.M.
        • Hay J.H.
        • Tsang V.H.
        • Durham J.S.
        A randomised trial of accelerated versus conventional radiotherapy in head and neck cancer.
        Radiother Oncol. 1997; 43: 39-46
        • Olmi P.
        • Crispino S.
        • Fallai C.
        • Torri V.
        • Rossi F.
        • Bolner A.
        • et al.
        Locoregionally advanced carcinoma of the oropharynx: conventional radiotherapy vs. accelerated hyperfractionated radiotherapy vs. concomitant radiotherapy and chemotherapy–a multicenter randomized trial.
        Int J Radiat Oncol Biol Phys. 2003; 55: 78-92
      1. Zackrisson B, Kjellén Elisabeth, Söderström K, Brun E, Nyman J NP. Five-year results from a Swedish study on conventional versus accelerated radiotherapy in head and neck squamous cell carcinoma – the ARTSCAN trial. Radiother Oncol 2015; Accepted for publication.

        • Hliniak A.
        • Gwiazdowska B.
        • Szutkowski Z.
        • Kraszewska E.
        • Kukolowicz P.
        • Jarzabski A.
        • et al.
        A multicentre randomized/controlled trial of a conventional versus modestly accelerated radiotherapy in the laryngeal cancer: influence of a 1 week shortening overall time.
        Radiother Oncol. 2002; 62: 1-10
        • Horiot J.C.
        • Bontemps P.
        • van den Bogaert W.
        • Le Fur R.
        • van den Weijngaert D.
        • Bolla M.
        • et al.
        Accelerated fractionation (AF) compared to conventional fractionation (CF) improves loco-regional control in the radiotherapy of advanced head and neck cancers: results of the EORTC 22851 randomized trial.
        Radiother Oncol. 1997; 44: 111-121
        • Beitler J.J.
        • Zhang Q.
        • Fu K.K.
        • Trotti A.
        • Spencer S.A.
        • Jones C.U.
        • et al.
        Final results of local-regional control and late toxicity of RTOG 9003: a randomized trial of altered fractionation radiation for locally advanced head and neck cancer.
        Int J Radiat Oncol Biol Phys. 2014; 89: 13-20
        • Skladowski K.
        • Maciejewski B.
        • Golen M.
        • Pilecki B.
        • Przeorek W.
        • Tarnawski R.
        Randomized clinical trial on 7-day-continuous accelerated irradiation (CAIR) of head and neck cancer – report on 3-year tumour control and normal tissue toxicity.
        Radiother Oncol. 2000; 55: 101-110
        • Bourhis J.
        • Overgaard J.
        • Audry H.
        • Ang K.K.
        • Saunders M.
        • Bernier J.
        • et al.
        Hyperfractionated or accelerated radiotherapy in head and neck cancer: a meta-analysis.
        Lancet. 2006; 368: 843-854
        • Al-Mamgani A.
        • van Rooij P.H.
        • Woutersen D.P.
        • Mehilal R.
        • Tans L.
        • Monserez D.
        • et al.
        Radiotherapy for T1–2N0 glottic cancer: a multivariate analysis of predictive factors for the long-term outcome in 1050 patients and a prospective assessment of quality of life and voice handicap index in a subset of 233 patients.
        Clin Otolaryngol. 2013; 38: 306-312
        • Yamazaki H.
        • Nishiyama K.
        • Tanaka E.
        • Koizumi M.
        • Chatani M.
        Radiotherapy for early glottic carcinoma (T1N0M0): results of prospective randomized study of radiation fraction size and overall treatment time.
        Int J Radiat Oncol Biol Phys. 2006; 64: 77-82
        • Moon S.H.
        • Cho K.H.
        • Chung E.J.
        • Lee C.G.
        • Lee K.C.
        • Chai G.Y.
        • et al.
        A prospective randomized trial comparing hypofractionation with conventional fractionation radiotherapy for T1–2 glottic squamous cell carcinomas: results of a Korean Radiation Oncology Group (KROG-0201) study.
        Radiother Oncol. 2014; 110: 98-103
        • Trotti A.
        • Zhang Q.
        • Bentzen S.M.
        • Emami B.
        • Hammond M.E.
        • Jones C.U.
        • et al.
        Randomized trial of hyperfractionation versus conventional fractionation in T2 squamous cell carcinoma of the vocal cord (RTOG 9512).
        Int J Radiat Oncol Biol Phys. 2014; 89: 958-963
        • Overgaard J.
        • Alsner J.
        • Eriksen J.
        • Horsman M.R.
        • Grau C.
        Importance of overall treatment time for the response to radiotherapy in patients with squamous cell carcinoma of the head and neck.
        Rays. 2000; 25: 313-319
        • Dische S.
        • Saunders M.
        • Barrett A.
        • Harvey A.
        • Gibson D.
        • Parmar M.
        A randomised multicentre trial of CHART versus conventional radiotherapy in head and neck cancer.
        Radiother Oncol. 1997; 44: 123-136
        • Overgaard J.
        • Hjelm-Hansen M.
        • Johansen L.V.
        • Andersen A.P.
        Comparison of conventional and split-course radiotherapy as primary treatment in carcinoma of the larynx.
        Acta Oncol. 1988; 27: 147-152
        • Blanchard P.
        • Baujat B.
        • Holostenco V.
        • Bourredjem A.
        • Baey C.
        • Bourhis J.
        • et al.
        Meta-analysis of chemotherapy in head and neck cancer (MACH-NC): a comprehensive analysis by tumour site.
        Radiother Oncol. 2011; 100: 33-40
        • Bentzen J.
        • Toustrup K.
        • Eriksen J.G.
        • Primdahl H.
        • Andersen L.J.
        • Overgaard J.
        Locally advanced Head and Neck cancer treated with accelerated radiotherapy, the hypoxic modifier nimorazole and weekly cisplatin. Results from the DAHANCA 18 phase II study.
        Acta Oncol. 2015; 54: 1001-1007
        • Overgaard J.
        • Hansen H.S.
        • Overgaard M.
        • Bastholt L.
        • Berthelsen A.
        • Specht L.
        • et al.
        A randomized double-blind phase III study of nimorazole as a hypoxic radiosensitizer of primary radiotherapy in supraglottic larynx and pharynx carcinoma. Results of the Danish Head and Neck Cancer Study (DAHANCA) Protocol 5–85.
        Radiother Oncol. 1998; 46: 135-146
        • Overgaard J.
        • Hansen H.S.
        • Andersen A.P.
        • Hjelm-Hansen M.
        • Jorgensen K.
        • Sandberg E.
        • et al.
        Misonidazole combined with split-course radiotherapy in the treatment of invasive carcinoma of larynx and pharynx: report from the DAHANCA 2 study.
        Int J Radiat Oncol Biol Phys. 1989; 16: 1065-1068
        • TD of VALCS Group
        Induction chemotherapy plus radiation compared with surgery plus radiation in patients with advanced laryngeal cancer.
        N Engl J Med. 1991; 324: 1685-1690
        • Groome P.A.
        • O’Sullivan B.
        • Irish J.C.
        • Rothwell D.M.
        • Math K.S.
        • Bissett R.J.
        • et al.
        Glottic cancer in Ontario, Canada and the SEER areas of the United States. Do different management philosophies produce different outcome profiles?.
        J Clin Epidemiol. 2001; 54: 301-315
        • Boje C.R.
        • Dalton S.O.
        • Gronborg T.K.
        • Primdahl H.
        • Kristensen C.A.
        • Andersen E.
        • et al.
        The impact of comorbidity on outcome in 12 623 Danish head and neck cancer patients: a population based study from the DAHANCA database.
        Acta Oncol. 2013; 52: 285-293
        • Jakobsen J.
        • Hansen O.
        • Jorgensen K.E.
        • Bastholt L.
        Lymph node metastases from laryngeal and pharyngeal carcinomas–calculation of burden of metastasis and its impact on prognosis.
        Acta Oncol. 1998; 37: 489-493
        • Johansen L.V.
        • Grau C.
        • Overgaard J.
        Glottic carcinoma–patterns of failure and salvage treatment after curative radiotherapy in 861 consecutive patients.
        Radiother Oncol. 2002; 63: 257-267
        • Hoff C.M.
        Importance of hemoglobin concentration and its modification for the outcome of head and neck cancer patients treated with radiotherapy.
        Acta Oncol. 2012; 51: 419-432
        • Overgaard J.
        • Mohanti B.K.
        • Begum N.
        • Ali R.
        • Agarwal J.P.
        • Kuddu M.
        • et al.
        Five versus six fractions of radiotherapy per week for squamous-cell carcinoma of the head and neck (IAEA-ACC study): a randomised, multicentre trial.
        Lancet Oncol. 2010; 11: 553-560
        • Trotti A.
        • Machtay M.
        RTOG 9003: Legacies of a Landmark Trial.
        Int J Radiat Oncol. 2014; 90: 253-254
        • Rosenthal D.I.
        • Fuller C.D.
        • Peters L.J.
        • Thames H.D.
        Final Report of Radiation Therapy Oncology Group Protocol 9003: provocative, but limited conclusions from exploratory analyses.
        Int J Radiat Oncol. 2015; 92: 715-717