European Archives of Oto-Rhino-Laryngology

, Volume 270, Issue 6, pp 1909–1915

Pretreatment performance status and nutrition are associated with early mortality of locally advanced head and neck cancer patients undergoing concurrent chemoradiation

Authors

  • Pei-Hung Chang
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Kun-Yun Yeh
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Jen-Seng Huang
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Chien-Hong Lai
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Tsung-Han Wu
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Yii-Jenq Lan
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Jason Chien-Sheng Tsai
    • Department of Radiation Oncology, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Eric Yen-Chao Chen
    • Department of Radiation Oncology, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
  • Shih-Wei Yang
    • Department of Otolaryngology, Head and Neck Surgery, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
    • Division of Hematology/Oncology, Department of Internal Medicine, Chang Gung Memorial Hospital, Keelung and Chang Gung UniversityCollege of Medicine
    • Cancer CenterChang Gung Memorial Hospital
Head and Neck

DOI: 10.1007/s00405-012-2290-2

Cite this article as:
Chang, P., Yeh, K., Huang, J. et al. Eur Arch Otorhinolaryngol (2013) 270: 1909. doi:10.1007/s00405-012-2290-2

Abstract

Unexpected fatal events in patients with head and neck cancers undergoing concurrent chemoradiation therapy are a clinical concern. Malnutrition, which is reported frequently in head and neck cancer patients, are associated with immunity derangement. The purpose of this study was to identify risk factors for early death of patients undergoing chemoradiation. We retrospectively analyzed the records of 194 stage III, IVA, and IVB head and neck cancer patients who were treated with chemoradiation between 2007 and 2009. We defined early death as death while receiving chemoradiation or within 60 days of treatment completion. Risk factors for early death were tested using univariate and multivariate analyses. Fourteen patients (7.2 %) experienced early death, 78.6 % of whom died of infection. Univariate analysis revealed significant correlations between early death and several pretreatment variables, including Eastern Cooperative Oncology Group performance status (PS) >1, hemoglobin <10 g/dL, albumin <3 g/dL, body mass index (BMI) <19 kg/m2, and peripheral blood total lymphocyte count <700/μL. Multivariate analysis showed that PS >1, BMI <19 kg/m2, and peripheral blood total lymphocyte count <700/μL were independent variables associated with early death. Poor performance status and malnutrition before chemoradiation independently predict early death in locally advanced head and neck cancer patients undergoing chemoradiation. Cautious management of head and neck cancer patients with these risk factors is required throughout chemoradiation period.

Keywords

Head and neck cancerConcurrent chemoradiationRisk factorsEarly deathMalnutrition

Introduction

Concurrent chemoradiotherapy, either as primary or adjuvant treatment for head and neck cancer, improves tumor control, survival rates, and chances of organ preservation [1]. However, its efficacy comes with substantial toxicity that has the potential to be severe, and is sometimes fatal. Head and neck cancer patients are at risk for nutritional deficiency [2]. Malnutrition damages immune functions and increases patient vulnerability to infection while receiving intensive and multimodal treatment. Death of patients during the course of treatment or soon after completion of treatment is a serious adverse event and remains a major concern for physicians when initiating therapy. We, therefore, conducted this study to identify risk factors, which may contribute to the early death of head and neck cancer patients undergoing concurrent chemoradiotherapy.

Materials and methods

Patient characteristics

Patients who had biopsy-proven stage III, IVA, or IVB (TNM stage reclassified according to American Joint Committee on Cancer staging system published in 2002) head and neck squamous cell carcinoma of the oral cavity, oropharynx, and hypopharynx and were treated with either primary or adjuvant chemoradiation therapy were eligible for this study. Patients with recurrent cancer, distant metastasis, or another concomitant active cancer were excluded from the study. From January 2007 to December 2009, we retrospectively reviewed the records of 201 patients with stage III, IVA, and IVB head and neck squamous cell carcinoma who had undergone chemoradiation therapy at our institution. Seven patients were excluded—three patients were lost to follow-up during chemoradiation period, two patients treated with re-chemoradiation for recurrent disease, and two patients were diagnosed with a concomitant active cancer upon diagnosis. A total of 194 patients who received concurrent chemoradiation were included in the analysis. All patients received intensity-modulated or arc technique radiotherapy on five consecutive days per week at a conventional fractionated daily dose of 1.8 or 2 Gy. The total prescribed dose of radiotherapy was 70–74 Gy for primary treatment and 60–66 Gy in the adjuvant setting. The initial treatment volume included the tumor bed and regional lymphatics. After receiving 46–50 Gy, the treatment area was reduced to irradiate the tumor bed and regional nodes. The chemotherapy regimens, including cisplatin 40 mg/m2 every 1 week, cisplatin 100 mg/m2 every 3 week, cisplatin 50 mg/m2 plus oral UFT capsule (tegafur plus uracil, 250 mg/m2/day) and oral calcium folinate (90 mg/day) on days 1–14 every 2 weeks, cetuximab 400 mg/m2 on cycle 1 only, and 250 mg/m2 since subsequent cycle every 1 week, were administered concurrently with radiotherapy according to the treatment guidelines at our institution. All patients were routinely referred to an early and intensive nutritional support program established in 2007 in our institution [3], which includes biweekly dietitian visits, mandatory feeding tube placement, timely calorie supplementation and blood transfusion as needed, continues on a regular basis throughout the chemoradiation period on an inpatient basis. More than 90 % of head and neck cancer patients undergoing concurrent chemoradiotherapy remain in hospital for completion of the treatment course. Our patients have government support healthcare via National Health Insurance Program in Taiwan. All patients were required to place the feeding tube placement including nasogastric tube or percutaneous gastrostomy if their body weight loss is more than 5 % during treatment course and to monitor closely the daily nutritional intake in the hospital throughout the treatment period to complement their oral intake to fulfill their energy requirement. This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital.

Statistical analysis

The major endpoint of this study was the death of the patient during the course of treatment or within the 60 days following completion of treatment, referred to as early death. For each case, we recorded age, gender, primary site of disease, stage at diagnosis, Eastern Cooperative Oncology Group performance status (PS), comorbid disease, body mass index (BMI), and biological parameters [serum albumin (Alb), peripheral blood lymphocyte counts (TLC), and hemoglobin level (Hb)]. Only those measurements taken fewer than 30 days prior to treatment initiation were included. Statistical analysis was performed using the SAS program, version 9.13 (SAS Institute, Cary, NC, USA). Risk factors for early death were tested using univariate and multivariate analyses. The difference among weight change and cisplatin accumulation dose between early death patients and non-early death patients during therapy were evaluated by independent Student t test. The univariate analysis was used to detect differences between the demographic characteristics of early death patients and non-early death patients. We used Pearson Chi-square (χ2) tests or Fisher’s exact test (when the expected number per cell <5) for this analysis. Multivariate analysis was performed using logistic regression with a forward regression procedure and a P value <0.05 required for entry. We used a 5 % significance level to determine independent factors for early death.

Results

Patients were predominately male (91.2 %) and the age at diagnosis ranged from 26 to 83 years (median 50 years). Fourteen patients (7.2 %) died during the course of treatment or within the 60 days following the completion of treatment (early death group). Among the 14 patients in the early death group, 11 died of severe infection (78.6 %), 1 died of rapid cancer progression, 1 died of tumor bleeding, and 1 died of ischemic colitis. In the early death group, nine patients received primary chemoradiation and five patients received adjuvant chemoradiation. In the non-early death group, 121 patients received primary chemoradiation and 59 patients received adjuvant chemoradiation. There was no significant difference between the early death group and the non-early death group in the following variables: age, gender, TNM stage, cancer location, Charlson comorbidity index, history of smoking, history of alcohol use, and betel quid exposure. However, several pretreatment variables, including PS >1, hemoglobin <10 g/dL, albumin <3 g/dL, BMI <19 kg/m2, and peripheral blood TLC <700/μL, were significantly correlated with the occurrence of early death as determined by univariate analysis (Table 1). Early death patients received a lower cisplatin dose than non-early death patients (134 ± 66.8 mg/m2 vs. 197.2 ± 64.8 mg/m2, P < 0.05). There was no difference about weight change during therapy between early death and non-early death group patients (−2.57 ± 5.33 kg vs. −3.12 ± 7.34 kg, P = 0.314). Multivariate analysis showed that BMI <19 kg/m2 (P = 0.049), peripheral blood TLC <700/μL (P = 0.038), and PS >1 (P = 0.001) were independently correlated with early death in head and neck cancer patients undergoing concurrent chemoradiation (Table 1). Clinical features of 14 early death patients are listed in Table 2.
Table 1

Demographic characteristics and statistical analysis for early and non-early death patients

 

Early death

Non-early death

Univariate analysis

Multivariate analysis

P value

Odds ratio (95 % CI)

P value

No. of patients

14

180

   

Age (years)

  

1.000

  

 ≥60

3

43

   

 <60

11

137

   

Gender

  

0.353

  

 Male

12

165

   

 Female

2

15

   

TNM stage at diagnosis

  

1.000

  

 III

1

20

   

 IVA, IVB

13

160

   

Cancer location

  

0.239

  

 Oral cavity

4

87

   

 Oropharynx

7

53

   

 Hypopharynx

3

40

   

Treatment modality

  

0.768

  

 CCRT

9

121

   

 Surgery + CCRT

5

59

   

Charlson comorbidity index

  

1.000

  

 ≤1

12

144

   

 >1

2

36

   

Exposure to smoking

  

0.473

  

 Yes

11

152

   

 No

3

28

   

Exposure to alcohol

  

0.788

  

 Yes

9

109

   

 No

5

71

   

Exposure to betel quid

  

0.221

  

 Yes

11

112

   

 No

3

68

   

ECOG performance status

  

<0.001

8.718 (2.550–29.809)

0.001

 ≤1

5

150

   

 >1

9

30

   

BMI (kg/m2)

  

0.034

3.865 (1.008–14.816)

0.049

 <19

5

23

   

 ≥19

9

157

   

Alb (g/dL)

  

0.037

  

 <3.0

5

23

   

 ≥3.0

9

154

   

Hb (g/dL)

  

0.050

  

 <10

4

10

   

 ≥10

10

163

   

Peripheral blood TLC (per μL)

  

0.01

4.507 (1.082–16.215)

0.038

 <700

5

16

   

 ≥700

9

164

   

Bold values indicate statistical significance

BMI body mass index, Alb albumin, Hb hemoglobin, TLC total lymphocyte count, CCRT concurrent chemoradiation

Table 2

Clinical features of 14 early death patients

Variables

Gender

Age

Primary site

Stage

Treatment modality

Operation

CT

RT

Complete treatment or interrupt

PS

BMI (kg/m2)

Alb (g/dL)

TLC (per μL)

Hb (g/dL)

CCI

Cause of death

1

Male

83

Oral cavity

III

OP + CCRT

Right hemiglossectomy + right neck LND

P + F

32 Gy/16fx

I

2

18.9

3.9

316

9.9

3

Ischemic bowel disease

2

Male

49

Hypopharynx

IVB

CCRT

 

P

30 Gy/15fx

I

2

22.6

2.8

491.4

8.2

6

Sepsis, pneumonia

3

Male

55

Oropharynx

IVA

CCRT

 

P + Cetuximab

34 Gy/17fx

I

0

16

4.5

640

12.8

0

Sepsis, neutropenic fever

4

Female

80

Oral cavity

IVA

CCRT

 

P

72 Gy/40fx

C

2

28.6

3

660

9.7

0

Sepsis, deep neck infection

5

Male

48

Oral cavity

IVA

OP + CCRT

Total glossectomy + right neck LND

P

48.2 Gy/25fx

I

1

15.1

3.6

609.6

13.2

0

Cancer progression

6

Female

47

Oral cavity

IVA

OP + CCRT

Left tongue tumor wide excision + left neck LND

P

54 Gy/30fx

I

2

19.6

3.4

824.1

9.1

1

Sepsis, neutropenic fever

7

Male

37

Hypopharynx

IVA

CCRT

 

P + U + L

32 Gy/16fx

I

1

22.8

3.8

998.4

13.5

0

Sepsis

8

Male

43

Oropharynx

IVA

CCRT

 

P + U + L

72 Gy/36fx

C

2

23.6

1.8

1,060.5

12.7

0

Sepsis, Neutropenic fever

9

Male

43

Oropharynx

IVB

CCRT

 

P

10.8 Gy/6fx

I

2

13.3

4

1,188

11.1

1

Sepsis, Neutropenic fever

10

Male

54

Oral cavity

IVA

OP + CCRT

Right buccal tumor wide excision + bilateral LND

P + F

54 Gy/27fx

I

1

22.5

3.2

1,361.2

12

0

Sepsis, empyema

11

Male

54

Hypopharynx

IVB

CCRT

 

P + U + L

66 Gy/33fx

C

2

24.3

1.9

1,576.8

12.3

1

Sepsis, neutropenic fever

12

Male

55

Oral cavity

IVA

OP + CCRT

Right tongue tumor wide excision + right neck LND

P

66 Gy/33fx

C

2

16.6

4.4

1,983.7

11.5

0

Sepsis, pneumonia

13

Male

60

Oropharynx

IVB

CCRT

 

P

38 Gy/19fx

I

1

19.6

2.1

2,662.2

12.2

1

Tumor bleeding

14

Male

46

Oropharynx

IVA

CCRT

 

P

75.6 Gy/42fx

C

3

19.4

2.5

2,841.6

11.6

1

Sepsis

CT chemotherapy, PS performance status, BMI body mass index, Alb albumin, TLC total lymphocyte count, Hb hemoglobin, CCI Charlson comorbidity index, OP operation, CCRT concurrent chemoradiation therapy, LND lymph node dissection, P cisplatin, F fluorouracil, U UFT, L leucovorin, I interrupt, C complete

Discussion

Concurrent chemoradiation therapy evolved as a definitive treatment for patients with locally advanced-stage head and neck cancer and has shown a significant survival benefit, higher laryngectomy-free survival rate, locoregional control rate, and lower distant metastasis rate than radiation alone [47]. Postoperative chemoradiation as an adjuvant treatment also offers benefits in locoregional control and survival for head and neck cancer patients with extracapsular spreading and positive surgical margin [810]. Although the chemoradiation benefits, such as increased survival, organ preservation, and locoregional control, are encouraging, the appropriate management of chemoradiation-related toxicity remains a clinical challenge during treatment. Mortality that occurs during treatment or soon after treatment has ended is a particularly serious adverse event, although the reported incidence rate is often <5 % [79, 1116] (Table 3). The risk factors for death during the course of treatment remain unknown. The present study examined several clinical variables and found that the rate of early death in head and neck cancer patients undergoing chemoradiation was 7.2 %. The pretreatment performance status and nutritional condition (BMI and TLC) independently predict early death.
Table 3

Reported toxic death rate of head and neck cancer patients undergoing chemoradiation

Trial

Role of CRT

Radiation therapy

Chemotherapy

Toxic death rate

RTOG 99-14 [11]

Primary CRT

72 Gy over 6 weeks; single arm, phase II

Cisplatin

CRT: 3.6 %

Staar et al. [12]

Primary CRT

69.9 Gy over 38 days

Fluorouracil + carboplatin

CRT arm: 1.8 %; RT arm: 2.4 %

RTOG 91-11 [7]

Primary CRT

70 Gy over 7 weeks

Cisplatin

CRT arm: 5 %; RT arm: 3 %; CT follow up by RT arm: 3 %

Intergroup 0126 [13]

Primary CRT

70 Gy over 7 weeks

Cisplatin

CRT arm: 4.2 %; CRT (split RT) arm: 2.1 %; RT arm: 2 %

Abitbol et al. [14]

Primary CRT

74.4 Gy over 16 weeks

Cisplatin; fluorouracil + MMC

CRT: 5 %

GORTEC 94-01 [15]

Primary CRT

70 Gy over 7 weeks

Carboplatin + fluorouracil

CRT arm: 0.9 %

RTOG 9501 [8]

Adjuvant CRT

60–66 Gy over 6–6.6 weeks

Cisplatin

CRT arm: 2 %; RT arm: 0 %

EORTC 22931 [9]

Adjuvant CRT

66 Gy over 6.5 weeks

Cisplatin

CRT arm: 0.6 %; RT arm: 0.6 %

TAX 324 [16]

Induction CT + Primary CRT

70–74 Gy over 7–7.4 weeks

Carboplatin

TPF induction + CRT arm: 0.4 %; PF induction + CRT arm: 0.4 %

CRT chemoradiation therapy, RT radiotherapy, CT chemotherapy, MMC mitomycin C, P cisplatin, F fluorouracil, T docetaxel

The early death rate observed in the current investigation was higher than those reported in previous studies. We believe this difference is due to the following factors: inclusion of patients with PS >1 and poor nutritional status in clinical practice. In our study, 39 of 194 patients (20.1 %) had PS >1, and many earlier studies excluded patients with performance status at that level. Further, the multivariate analysis results supported PS >1 as an independent risk factor for early death. This result implies that concurrent chemoradiotherapy may not be suitable for patients with poor performance status.

Additionally, head and neck cancer patients are at risk for nutritional deficiency [17]. Malnutrition alters defense mechanisms, including anatomic barriers, cell-mediated immune responses, phagocytes/microbicidal functions, and humoral immunity (antibody and complement responses) [1820]. Severely malnourished patients are often characterized by functional complement deficit, diminished opsonic serum capacity, diminished antibody response avidity, inhibition of neutrophil and macrophage migration, reduced phagocyte intracellular killing capacity, decreased secretory IgA production, and reduced ability to alter endocrine function [21, 22]. Severe malnutrition has a negative impact on both short-term mortality and overall survival in head and neck cancer [23]. Multiple factors undermine the nutritional status of head and neck patients, such as a pre-morbid lifestyle with poor dietary habits that is often combined with excessive smoking and alcohol consumption [24]. In addition to the alterations in chewing and swallowing that can accompany specific tumors or their surgical treatment, aggressive chemotherapy and radiotherapy can lead to dysphagia, odynophagia, and dehydration [25], all of which may result in malnutrition, treatment interruption, morbidity, and even mortality [26]. The univariate analysis in the present study showed that the nutrition indicators, including low BMI (BMI <19 kg/m2), serum albumin (<3 mg/dL), total blood lymphocyte count (<700/μL), and hemoglobin level (<10 g/dL) were correlated with early death. Weight loss or low BMI in head and neck cancer patients increase morbidity and mortality and decreases treatment tolerance and overall quality of life [23, 2729]. TLC represents patient nutrition status and immunity. Severe lymphopenia is a predictor of chemotherapy complications and a prognostic factor in several clinical situations [3035]. The majority of early death patients (78.6 %) in our study died of severe infection. The multivariate analysis identified BMI <19 kg/m2 and TLC <700/μL as independent risk factors for early death, suggesting that optimizing nutrition status is important for head and neck cancer patients undergoing concurrent chemoradiation.

Our study result is analyzed from real-world daily clinical practice where heterogeneous treatment patterns and patient populations may lead to different estimates than those observed in clinical trials. The present study pointed that two over 80-year-old patients (No. 1 and No. 4 in Table 2) suffered early mortality during the treatment course. This observation may not be commonly observed in the clinical trials that usually excluded patients with extremely old age. Although older patients have higher rates of hematologic toxicity and treatment-related death than younger during adjuvant chemotherapy [36], clinicians may treat patients with older age under consideration of performance status, treatment modification, disease control and patients’ expectation. The current two cases suggest that less aggressive treatment schedule, more intensive supportive care and comprehensive communication between physicians and patients before treatment initiation may be required for these older patients who intend to receive chemoradiotherapy for survival benefit.

In conclusion, we identified poor performance status, low BMI, and severe lymphopenia as independent risk factors for early death in head and neck cancer patients undergoing concurrent chemoradiation. An experienced multidisciplinary team should optimize both the performance status and nutritional condition of locally advanced head and neck cancer patients before the initiation of concurrent chemoradiation treatment in order to avoid the tragedy of early death. Less aggressive treatment schedule such as chemotherapy dosage modification or radiotherapy only should be considered in patients with these risk factors mentioned above. Further randomized prospective study is warranted to examine these observations.

Acknowledgments

The authors thank all the members of the Cancer center, Chang Gung Memorial Hospital, Keelung, for their invaluable help. The authors would also like to thank Wei-Jhih Wang for the assistance in the data collection and statistics analysis.

Conflict of interest

The authors declare that they have no competing interests and no financial relationship with other organizations sponsoring this research. All authors have nothing to disclose.

Copyright information

© Springer-Verlag Berlin Heidelberg 2012