Annals of Surgical Oncology

, Volume 16, Issue 5, pp 1337–1343

Feasibility and Efficacy of Accelerated Weekly Concomitant Boost Postoperative Radiation Therapy Combined with Concomitant Chemotherapy in Patients with Locally Advanced Head and Neck Cancer

Authors

  • Berrin Pehlivan
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Francois Luthi
    • Department of Medical OncologyUniversity Hospital Center, University of Lausanne
  • Oscar Matzinger
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Michael Betz
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Daniela Dragusanu
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Shelley Bulling
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Luc Bron
    • Department of OtorhinolaryngologyUniversity Hospital Center, University of Lausanne
  • Philippe Pasche
    • Department of OtorhinolaryngologyUniversity Hospital Center, University of Lausanne
  • Walter Seelentag
    • Department of PathologyUniversity Hospital Center, University of Lausanne
  • René O. Mirimanoff
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
  • Abderrahim Zouhair
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
    • Department of Radiation OncologyUniversity Hospital Center, University of Lausanne
Head and Neck Oncology

DOI: 10.1245/s10434-009-0426-4

Cite this article as:
Pehlivan, B., Luthi, F., Matzinger, O. et al. Ann Surg Oncol (2009) 16: 1337. doi:10.1245/s10434-009-0426-4

Background

The aim of this study was to assess feasibility and efficacy of weekly concomitant boost accelerated postoperative radiation therapy (PORT) with concomitant chemotherapy (CT) in patients with locally advanced head and neck cancer (LAHNC).

Methods and Materials

Conformal or intensity-modulated 66-Gy RT was performed in 5.5 weeks in 40 patients. Cisplatin was given at days 1, 22, and 43. Median follow-up was 36 months.

Results and Discussion

Grade 3 mucositis, dysphagia, and erythema was observed in ten (25%), nine (23%), and six (13%) patients, respectively. Grade 3 or more anemia was observed in two (6%) patients, and leukopenia in five (13%) patients. No grade 3 or 4 thrombocytopenia was observed. Grade 3 nephrotoxicity was observed in one patient (3%). No treatment-related mortality was observed. Grade 2 or more xerostomia and edema were observed in ten (25%) and one (3%) patient, respectively. Locoregional relapse occurred in eight patients, and seven patients developed distant metastases. Median time to locoregional relapse was 6 months. Three-year overall, disease-free survival, and locoregional control rates were 63%, 62%, and 81%, respectively. Multivariate analysis revealed that the only prognostic factor was nodal status.

Conclusion

Reducing overall treatment time using accelerated PORT/CT by weekly concomitant boost (six fractions per week) combined with concomitant cisplatin CT is easily feasible with acceptable morbidity.

Following surgery, postoperative radiotherapy (RT) combined with concomitant chemotherapy (CT) improves the outcome in locally advanced head and neck cancer (LAHNC).1,2 Despite multimodality treatment, locoregional relapse risk remains as high as 35–40%.35 The risk of recurrence or metastasis is particularly high in patients with inadequate resection margins, extranodal spread or multiple involved lymph nodes.6 Various strategies have been proposed to improve the outcome.710 Two major approaches have gained importance in recent years to improve the outcomes: altered fractionation and concomitant CT regimens. Confirming expectations based on radiobiological theory, altered fractionation regimens could overcome accelerated tumor population, and concurrent delivery of CT and ionizing radiation could sensitize tumor clonogens.

Data concerning hyperfractionated or accelerated RT with concomitant CT are very limited, and there are no prospective data in the postoperative setting. We previously showed the feasibility of accelerated (six fractions/week) fractionated 66-Gy postoperativer RT (PORT) using a single-fraction regimen from Monday to Thursday and a concomitant boost (two fractions separated by 6-h interval) at Friday afternoon sessions with acceptable toxicity.11 On the basis of these findings, in this study we combined concomitant cisplatin CT to accelerated PORT in patients with LAHNC in order to increase locoregional control (LRC) and survival.

Methods and Materials

Forty consecutive patients with advanced-stage squamous-cell carcinoma of head and neck treated at the University Hospital of Lausanne (CHUV) were included in this study. All patients were treated with curative surgery followed by accelerated PORT and concomitant CT. Inclusion criteria consisted of nonmetastatic LAHNC classified as pT1–pT4 and/or pN0–pN3, age 18 years or older, no previous history of cancer other than nonmelanoma skin cancer or in situ cervix cancer, and good performance status [World Health Organization (WHO) scale 0–1]. The study was approved by the Ethical Committee of the University of Lausanne. T- and N-classification were assigned according to the staging system of the International Union against Cancer (UICC) 2002.12 Pretreatment characteristics are listed in Table 1.
Table 1

Patient characteristics

 

Number

%

Tumor site

Oral cavity

15

38

Oropharynx

8

20

Hypopharynx

8

20

Larynx

7

18

Unknown primary

2

5

RT indications

Extranodal infiltration:

24

60

With (+) surgical margin

13

32

Without (+) surgical margin

11

28

Without extranodal infiltration

16

40

(+) surgical margin

9

22

T4 tumors

5

13

≥3 lymph nodes

2

5

pT-classification

0

2

5

1

5

12

2

16

40

3

2

5

4

15

38

pN-classification

0

3

7

1

8

20

2a

2

5

2b

15

38

2c

5

12

3

7

18

Pretreatment evaluation included a medical history, examination with panendoscopy (oro-pharyngo-laryngoscopy, bronchoscopy, and esophagoscopy), and computed tomography and/or magnetic resonance imaging (MRI) of the head and neck region in all cases. Additional diagnostic procedures for distant metastases, including computed tomography of the chest, liver ultrasound, bone scintigraphy, and/or more recently positron emission tomography, were only performed if clinically indicated. Laboratory studies included blood chemistry (electrolytes, liver and kidney function tests), and complete blood count were performed each week of the treatment.

All patients were seen and discussed at our multidisciplinary tumor board, and the treatment recommendation was made jointly by head and neck surgeons, radiation oncologists, and medical oncologists with advice from the diagnostic radiologist, pathologist, nuclear medicine specialist, nursing staff, nutritionists, and dentists. Percutaneous endoscopic gastrostomy (PEG) was placed before or at the beginning of treatment in 17 of the 40 (43%) patients, and 4 patients (10%) had nasogastric feeding tube on the basis of our previous experience.13 Nutritional intervention, professional surveillance, advice, and support were available during the entire treatment period; prophylactic mouth washes were recommended, and fungal or bacterial infections on mucous membranes were treated according to microbiologic testing. All patients had dental examination and, if necessary, dental treatment before start of treatment.

Postoperative Treatment

Postoperative accelerated 66-Gy RT was given using six fractions per week (2 Gy/fraction) with a single-fraction regimen from Monday to Thursday and a two fractions including a concomitant boost on Friday (with an interval of at least 6 h) during 5.5 weeks (38 days). RT was implemented in all patients using five-field three-dimensional (3D) conformal or intensity-modulated RT (IMRT), according to the Groupe Oncologie-Radiothérapie Tête et Cou (GORTEC)/European Organisation for Research and Treatment of Cancer (EORTC)/Radiation Therapy Oncology Group (RTOG) guidelines. The boost volume (66 Gy) consisted of the clinical target volume (CTV1), which was defined according to the presurgical location of the primary tumor, and its nodal extension including the entire surgical bed. Planning target volume (PTV1) included a 5-mm margin around the CTV1 in three dimensions. The initial treatment volume (PTV2, 46–50 Gy) consisted of the boost volume and all nodal areas at risk of subclinical malignant disease (PTV2), including a 5-mm margin around the CTV2 in three dimensions. Patient immobilization was realized using individualized thermoplastic immobilization masks in all patients.

Concomitant cisplatin CT was planned at 100 mg/m2 at days 1, 22, and 43 in all but five patients, in whom carboplatin [area under the curve (AUC) = 2 weekly, or AUC = 5 days 1, 22, and 43] was chosen due to inadequate renal function.

Patients received prophylactic hydration and antiemetic agents and were seen weekly by the radiation oncologist, head and neck surgeon, and/or medical oncologist during RT. Special advice and attention were given by the nursing team to skin care, and oral hygiene was enforced. Symptomatic treatment for mucositis (sodium bicarbonate) was offered, and oral candidiasis was promptly treated by fluconazol.

Toxicity Assessment and Follow-Up

Toxicity assesment was done according to the Common Toxicity Criteria (CTC)/National Cancer Institute (NCI) v3.0. Patients were not assessed by the same observer. Clinical examination and toxicity assessment was repeated weekly during RT. Patients were then followed by the head and neck surgeon and radiation oncologist weekly during the first 2 months, monthly assessments from 2–6 months, and every 2–3 months until 2 years. At that time, a panendoscopy was performed, and MRI, computed tomography, and biopsies were performed as indicated. Patients without evidence of disease were subsequently followed every 6 months up to 5 years, then once per year.

Statistical Methods

Cause-specific survival (CSS), disease-free survival (DFS), and overall survival (OS) rates as well as actuarial LRC rates were calculated using product-limit method.14 Time to any event was measured from the date of pathological diagnosis. The events were death (all causes) for OS, head and neck cancer-related mortality for CSS, and death (all causes) or relapse for DFS. For the LRC rate, the event consisted of local or regional relapse. Confidence intervals (CI) were calculated from standard errors. Differences between groups were assessed using the log-rank test.15 Multivariate analyses were done using the Cox stepwise-regression analysis to determine the independent contribution of each prognostic factor.16

Results

Between March 2001 and April 2006, 36 men and 4 women with pT1–pT4 and/or pN0–pN3 LAHNC were included in this study. Median age was 60 years (range 36–81 years). Indications of PORT/CT were positive surgical margins (n = 9; all R1), T4 R0 tumors (n = 5), three or more positive lymph nodes without extranodal infiltration (all R0; = 2) in 16 (40%) patients; or extranodal infiltration with (all R1; n = 13) or without (n = 11) positive surgical margins in 24 (60%) patients. Twenty-three patients with T0 (n = 2), T1 (n = 5), and T2 (n = 16) were included because of positive surgical margins (n = 4; all R1) or because of advanced nodal disease.

Median interval between surgery and RT was 46 days (range 24–112 days), and median RT duration was 39 days (range 35–62 days). In five patients, treatment duration was longer than 45 days because of toxicity. All but two patients received the planned total dose without unplanned interruption (66 Gy in 38, 64 Gy in 1, and 58 Gy in 1). Three cycles of planned full-dose cisplatin could be administered in 18 patients (54%). Because of toxicity, 15 patients (43%) could receive only two cycles of CT, and 2 patients only (3%) one cycle of CT. Dose reduction of cisplatin was necessary in two patients, who could receive only two cycles of CT. Carboplatin was given weekly in one patient, and every 3 weeks in four patients.

Three patients had local relapse only, one patient locoregional relapse, two patients regional relapse alone, two patients regional relapse and distant metastases, and five patients had only distant metastases (Fig. 1). Median time to locoregional relapse was 6 months (range 1–40 months). Three-year OS, CSS, DFS, and LRC rates were 63% (95% CI 46–78%), 66% (95% CI 50–82%), 62% (95% CI 48–78%), and 81% (95% CI: 69–93%), respectively (Fig. 2). Distant metastasis probability at 3 years was 20% (95% CI 7–33%). For univariate analyses gender, age, tumor site, pT- classification, pN-classification, extracapsular nodal extension, positive surgical margins, interval between surgery and RT, and total RT time were analyzed (Table 2). Univariate and multivariate analyses revealed that the only prognostic factor influencing the outcome was nodal status, for both DFS and LRC [relative risk (RR) = 0.40, P = 0.02; and RR = 0.0004, P = 0.02].
https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0426-4/MediaObjects/10434_2009_426_Fig1_HTML.gif
Fig. 1

Distribution of relapses in 40 patients treated with accelerated weekly concomitant boost postoperative radiation therapy combined with concomitant chemotherapy

https://static-content.springer.com/image/art%3A10.1245%2Fs10434-009-0426-4/MediaObjects/10434_2009_426_Fig2_HTML.gif
Fig. 2

Probability of overall survival (solid line), disease-free survival (dotted line), and locoregional failure (dashed line) in 40 patients treated with accelerated weekly concomitant boost postoperative radiation therapy combined with concomitant chemotherapy

Table 2

Univariate analyses

 

N

3-year

95% CI

P-value

3-year

95% CI

P-value

DFS

LRC

All patients

40

62

48–78

81

69–93

Gender

Female

4

100

0.13

100

0.29

Male

36

59

43–75

79

65–93

Age (years)

<60

20

74

54–94

0.35

90

76–100

0.47

≥60

20

53

31–75

73

53–93

Tumor site

Oral cavity

15

63

38–87

0.97

78

56–100

0.95

Oropharynx

8

63

30–99

75

45–100

Larynx

7

71

38–104

86

61–100

Hypopharynx

8

51

11–91

86

61–100

Unknown primary

2

T-classification

pT1-2

23

65

46–84

0.85

78

60–96

0.51

pT3-4

17

61

46–86

88

72–100

N-classification

pN0-1

11

88

64–112

0.04

100

0.07

pN2-3

29

55

37–73

75

59–91

ENE

No

16

66

44–96

0.81

79

68–90

0.96

Yes

24

61

41–79

 

83

67–99

Positive surgical margin

No

18

64

41–97

0.92

76

54–98

0.34

Yes

22

62

41–95

85

69–100

Surgery–RT interval

≤42 days

16

56

32–80

0.47

75

53–97

0.56

>42 days

24

68

48–88

86

70–100

RT duration

≤39 days

21

61

39–83

0.62

80

62–98

0.88

>39 days

19

66

44–88

84

66–100

ENE extranodal infiltration

Incidence of grade 3 mucositis was 25% (n = 10), grade 3 dysphagia 23% (n = 9), and grade 3 skin erythema 13% (n = 5). CT-related anemia was observed in two patients (grade 3 in one, and grade 4 in one), leukopenia in four patients (grade 3 in two, and grade 4 in two), and no grade 3 or 4 thrombopenia was observed. Grade 3 renal toxicity was observed in only one patient (Table 3). Median weight loss was 3.5 kg (range 0–14.5 kg). No treatment-related mortality was observed.
Table 3

Acute toxicity

Toxicity

Grade I (%)

Grade II (%)

Grade III (%)

Grade IV (%)

Mucositis

11 (28)

19 (47)

10 (25)

Dysphagia

9 (23)

22 (54)

9 (23)

Erythema

13 (32)

21 (53)

6 (15)

Anemia

27 (68)

7 (18)

1 (3)

1 (3)

Leukopenia

7 (18)

3 (8)

2 (5)

2 (5)

Thrombocytopenia

12 (30)

Renal toxicity

10 (25)

2 (5)

1 (3)

Considering late effects, grade 0, 1, 2 or 3 xerostomia was observed in 9 (23%), 22 (55%), 9 (23%), and 1 (3%) patient, respectively; grade 0, 1, and 2 edema was observed in 25 (63%), 14 (35%), and 1 (3%) patient, respectively. Two patients developed mandibular osteoradionecrosis.

Discussion

The impact of shortening overall treatment time on therapeutic outcome is well established. Several studies validated that, compared with conventional fractionation using 2 Gy per day, altered fractionation (especially hyperfractionated or accelerated RT) increases locoregional tumor control in patients with HNSCC.9,1719 Ang et al. compared an accelerated concomitant boost PORT delivering a total dose of 63 Gy in 35 fractions of 1.8 Gy during 5 weeks (five fractions/week during 3 weeks and two fractions/day during 2 weeks) with a conventionally fractionated PORT delivering a total dose of 63 Gy in 35 fractions of 1.8 Gy/fraction during 7 weeks.20 This study, in the accelerated PORT arm, showed higher but not statistically significant improvement in 5-year LRC (P = 0.11) and OS (P = 0.08) in favor of accelerated arm. However, it confirmed prospectively that, in patients treated by conventional fractionation, delay of more than 6 weeks between surgery and RT significantly reduces LRC and OS. Another important study by Overgaard et al. compared a modestly accelerated regimen of 66–70 Gy in 2-Gy fractions given six times per week with the same dose given in five fractions per week, using nimorazole in both groups.21 They reported a gain of 15% in local control using accelerated RT, with increased acute toxicity but no increase in late toxicity. In the postoperative setting, Sanguineti et al. and Swunski et al. confirmed the feasibility of accelerated PORT.22,23 Like Ang et al., Sanguineti et al. found that patients delaying the start of RT benefited more from accelerated RT in terms of LRC; however, they did not observe any difference between the standard and accelerated treatment arms (2-year LRC of 80% and 78%, respectively). Swunski et al. reported acceptable acute toxicity, in spite of high intensity and duration of mucosal reactions, in the continous accelerated treatment arm.

The impact of overall treatment time on outcome is more important for locoregional control than survival. Overall, the incidence of distant metastases can reach 20–40%. Several studies have demonstrated that the addition of CT to RT improves survival. The EORTC and the RTOG have both recently conducted randomized phase III trials comparing this approach with standard PORT.1,2 The EORTC 22931 study showed improved LRC, DFS, and OS with the combined treatment modality. However, the RTOG 10-95 study failed to show a significant difference in terms of OS whereas LRC and DFS were significantly improved by the combined modality. A combined analysis of both studies revealed that addition of concomitant chemotherapy to PORT results in significantly better outcome, especially in patients with inadequate resection margins and/or extranodal spread.6

Concomitant CT and hyperfractionated or accelerated RT without surgery has been confirmed by several studies; however, there are no data in the postoperative setting.2428 In our department, accelerated RT has been implemented since 1998. In our experience, concurrent administration of three cycles of full-dose cisplatin and 5-fluorouracil (5-FU) CT and concomitant boost accelerated 70-Gy RT is feasible in patients with advanced squamous cell carcinoma of the head and neck.29 We also reported our results using accelerated PORT, in which 2-year actuarial LRC was 80%, and distant metastases probabilty was 38% with acceptable toxicity.11 In the present study, we aimed to assess whether weekly concomitant boost (six fractions per week) accelerated PORT combined with cisplatin CT is better than accelerated PORT alone. Three-year OS and LRC rates were 65% and 82%, respectively; distant metastases probability at 3 years was 19%. The incidence of grade 3 mucositis was 25%, grade 3 dysphagia 23%, and grade 3 skin erythema 13%. CT-related bone marrow toxicity was observed in six patients (15%), and grade 3 renal function impairment was observed in only one patient. In this study, 95% of patients received the prescribed full dose of radiation, and only 54% of patients were able to complete the full cisplatin dose (three cycles) with moderate toxicity, a finding also observed in the EORTC and RTOG studies.1,2 In the future, it should be considered whether or not to continue with this third cycle of cisplatin. Acceptable toxicity rates in our series are most likely due to our relatively soft accelerated RT schedule. Treating the boost volume every Friday afternoon during the first 5 weeks by using smaller PTVs is easily accepted by patients as well as the radiotherapy department.

The results of this trial demonstrated that reducing the overall treatment time using accelerated PORT/CT by weekly concomitant boost (six fractions per week) combined with concomitant cisplatin CT is easily feasible with good locoregional and distant metastases control in curatively operated patients with LAHNC. However, acute toxicity remains the main obstacle to wide acceptance of this treatment modality requiring close follow-up and supportive care by an experienced multidisciplinary team. Current research strategies will continue to focus on improving the therapeutic ratio in an attempt to achieve higher rates of both local and systemic control. New targeted therapy agents with less toxicity should be investigated, combined with current chemoradiation regimens.

Copyright information

© Society of Surgical Oncology 2009