Int J Clin Oncol
DOI 10.1007/s10147-017-1201-3
ORIGINAL ARTICLE
Esophageal squamous cell neoplasia is an independent negative
prognostic factor for head and neck cancer patients
Yasuhiko Hamada1 · Toshiro Mizuno2 · Kyosuke Tanaka1 · Masaki Katsurahara1 ·
Noriyuki Horiki1 · Reiko Yamada3 · Hiroyuki Inoue3 · Yoshiyuki Takei3 ·
Naoyuki Katayama2 
Received: 30 April 2017 / Accepted: 7 October 2017
© Japan Society of Clinical Oncology 2017
Abstract  Conclusions  Our study suggests that although the survival
Background  Patients with head and neck cancer (HNC)
have a high incidence of esophageal squamous cell neo-
plasms (ESCN). ESCN also has a negative impact on the
survival of HNC patients. However, recent endoscopic
advances enable the early detection of ESCN, and novel
treatments may lead to improving survival rates for HNC
patients with ESCN.
Methods  HNC patients who underwent magnifying esoph-
agogastroduodenoscopy (EGDS) from 2005 to 2012 were
included in this study (n = 226). We analyzed the prevalence
and prognostic value of ESCN in HNC patients and the dif-
ference in overall survival between HNC patients with and
without ESCN.
Results  Thirty-four patients (15%) developed an ESCN
during their clinical course. Of the 34 patients, 10 patients
underwent endoscopic resection for ESCN and 10 patients
underwent simultaneous chemoradiation therapy for HNC
and ESCN. The 3-year survival rates in HNC patients with
and without ESCN were 53% and 70%, respectively. Multi-
variate analysis identified the advanced clinical stage of the
HNC [hazard ratio (HR) = 2.15; 95% confidence interval
(CI) = 1.18–3.93; p = 0.012] and the presence of ESCN (HR
= 1.73; 95% CI = 1.00–2.97; p = 0.049) as significant and
independent determinants of overall survival.
* Yasuhiko Hamada
y‑hamada@clin.medic.mie‑u.ac.jp
1
Department of Endoscopic Medicine, Mie University
Hospital, 2‑174 Edobashi, Tsu, Mie 514‑8507, Japan
2
Department of Hematology and Oncology, Mie University
Graduate School of Medicine, Tsu, Japan
3
Department of Gastroenterology and Hepatology, Mie
University Graduate School of Medicine, Tsu, Japan
of HNC patients with ESCN may be improved by routine
EGDS during tumor surveys and by advances in endoscopy,
the presence of ESCN still remains an independent negative
prognostic factor for HNC patients.
Keywords  Head and neck cancer · Esophageal
neoplasia · Prognostic factor
Introduction
Head and neck cancer (HNC) is the seventh most common
cancer worldwide, with more than 600,000 patients diag-
nosed with HNC annually [1]. HNC is associated with a
high likelihood of second primary malignancies. The most
common site for this type of cancer is an aerodigestive site,
including the lungs and esophagus [2, 3]. Epidemiological
studies have suggested an increased risk for the development
of HNC is associated with smoking and alcohol drinking
habits; these risk factors overlap those related to such aer-
odigestive malignancies [4, 5].
Patients with HNC have shown a high prevalence of
esophageal squamous cell neoplasia (ESCN). Specifically,
from 5% to 15% of patients with HNC ultimately develop
synchronous or metachronous ESCN [6–8]. A screening
esophagogastroduodenoscopy (EGDS) to conduct a tumor
survey is routinely recommended for patients with HNC
[9–11].
Moreover, ESCN was found to have a significantly nega-
tive impact on the survival of HNC patients. Studies have
shown the 3-year overall survival rate to be between 0% and
15% in HNC patients [12–14]. However, recent advances
in endoscopy have enabled the early detection of esopha-
geal neoplasia [8]. Therefore, in HNC patients today, the
13
Vol.:(0123456789)

incidence of advanced ESCN may be lower and the present
overall survival rate may have improved from those of past
reports.
The objective of this study was to investigate the preva-
lence of ESCN and the impact on the overall survival rate of
HNC patients in recent years. We retrospectively examined
such questions using the medical data of the HNC patients
who had undergone EGDS at our institute.
Patients and methods
Patients
We retrospectively reviewed the medical records of HCN
patients who had undergone EGDS at our institute between
October 2005 and September 2012. Patients were selected
for this study according to the following criteria:
i. At the time of an HCN diagnosis, patients underwent
staging workups, including oral cavity inspection,
endoscopic examination of the pharynx and larynx,
computed tomography (CT) and/or magnetic reso-
nance imaging of the head and neck, and whole-body
fluorodeoxyglucose–positron emission tomography.
ii. For those patients with HCN who underwent EGDS,
EGDS were performed with an upper gastrointestinal
high-resolution magnifying endoscope (GIF-Q240Z,
GIF-H260Z; Olympus Medical Systems, Tokyo,
Japan).
iii. After treatment for HCN, CT and EGDS were per-
formed annually to check for the recurrence of disease
and to survey for ESCN.
Study design
We analyzed the prevalence of ESCN in HNC patients, its
association with patient baseline characteristics (sex, age,
toxic habits such as smoking and alcohol drinking, loca-
tion of HNC, staging of HNC, location of second primary
malignancies, initial treatment for HNC), the difference in
therapeutic outcomes of HNC patients with and without
ESCN, and the prognostic value of the presence of ESCN
in HNC patients.
The data of HNC patients with ESCN were divided into
two groups, depending on whether patients had synchronous
or metachronous ESCN. Synchronous ESCN was defined
as developing at the same time, or within 6 months before
and after a diagnosis of HNC. For development before and
after 6 months, the ESCN was considered metachronous. In
cases of HNC with multiple ESCNs, the ESCN stage was
defined according to the most advanced lesion. Alcohol
13
Int J Clin Oncol
consumption was estimated as the average number of grams
of pure ethanol ingested per drinking day. Smoking was cal-
culated according to the Brinkman index (the number of
cigarettes smoked per day multiplied by the number of total
smoking years). The primary tumor was grouped by location
(oral cavity, epipharynx, oropharynx, hypopharynx, larynx).
Tumor stage was classified according to TNM classification
of the Union for International Cancer Control, 6th edition
[15].
Statistical analysis
Continuous variables were described by medians and ranges;
categorical variables were described by counts and percent-
ages. All patient data were compared based on whether
patients had ESCN. Continuous variables were compared
using the Wilcoxon rank-sum test. Categorical variables
were compared with the chi-square test when appropriate;
otherwise, a Fisher’s exact test was used. Cumulative inci-
dences of overall survival of HNC patients, with or without
ESCN, were calculated by the Kaplan–Meier method, with
the date of the first EGDS at our institute as the starting
point, and differences were assessed with the log-rank test.
The prognostic factors that were shown to be significant in
a univariate Cox regression analysis were also tested with a
multivariate Cox regression analysis. We used the analysis
to examine hazard ratio (HR) of the HNC patients, with the
95% confidence interval (CI). We conducted all statistical
analyses using SPSS version 20 (SPSS, Chicago, IL, USA).
All statistical tests were two tailed, and significance was
defined as p < 0.05. This study was approved by the institu-
tional review board of Mie University Hospital.
Results
Baseline characteristics of patients with HNC
During the study period, a total of 226 patients fulfilled
the eligibility criteria for this study (Table 1). Of the 226
patients, 194 (85.8%) were male; the median age was 68
years (range, 33–89 years). The median amount of daily
alcohol intake was 22 g (range, 0–220 g) and the median
Brinkman index was 606 (range, 0–2560). Considering
the location of the HNC, 40 lesions (17.7%) were found
in the oral cavity, 12 lesions (5.3%) in the epipharynx, 42
lesions (18.6%) in the oropharynx, 74 lesions (32.7%) in the
hypopharynx, and 58 lesions (25.7%) in the larynx. Sixty-
seven patients (29.6%) had localized disease (stage I, II),
and 159 patients (70.4%) had locally advanced or metastatic
disease (stage III, IV). Twenty-three patients (10.2%) had
second primary malignancies other than ESCN (gastric
Int J Clin Oncol

Table 1  Patient background and comparison between those with and without ESCN
Characteristics Total (n = 226) Without ESCN (n = 192) With ESCN (n = 34) p*

Sex (male), n (%) 194 (85.8) 161 (83.9) 33 (97.1) 0.06

Age, median, year (range) 68 (33–89) 68 (33–89) 68 (51–88) 0.98
Alcohol intake, median, g/day (range)a 22 (0–220) 20 (0–220) 44 (0–220) 0.007
Brinkman index, median (range) 606 (0–2560) 520 (0–2560) 880 (0–1480) 0.26
Location of primary tumor, n (%)
 Oral cavity 40 (17.7) 38 (19.8) 2 (5.9) 0.053
 Epipharynx 12 (5.3) 12 (6.3) 0 (0.0) 0.22
 Oropharynx 42 (18.6) 33 (17.2) 9 (26.4) 0.23
 Hypopharynx 74 (32.7) 55 (28.6) 19 (55.9) 0.003
 Larynx 58 (25.7) 54 (28.1) 4 (11.8) 0.050
Clinical stage, n (%)
 Stage I 24 (10.6) 19 (9.9) 5 (14.7) 0.37
 Stage II 43 (19.0) 38 (19.8) 5 (14.7) 0.64
 Stage III 36 (15.9) 32 (16.7) 3 (8.8) 0.31
 Stage IV 123 (54.5) 103 (53.6) 21 (61.8) 0.38
 Second primary tumor except ESCN, n (%) 23 (10.2) 19 (9.9) 4 (11.8) 0.76
Initial treatment for HNC
 Surgery 73 (32.3) 68 (35.4) 5 (14.7) 0.030
 Radiotherapy 21 (9.3) 17 (8.9) 4 (11.8) 0.53
 Chemoradiation therapy 104 (46.0) 83 (43.2) 21 (61.7) 0.061
 Chemotherapy 18 (8.0) 16 (8.3) 2 (5.9) 1.00
 Other 10 (4.4) 8 (4.2) 2 (5.9) 0.65

ESCN esophageal squamous cell neoplasia, HNC head and neck cancer
* Chi-square or Fisher’s exact tests were used for sex, site of primary tumor, clinical stage, second primary tumor except ESCN, and initial treat-
ment for HNC; Wilcoxon rank-sum tests were used for age, alcohol intake, and Brinkman index
a
  Amount of alcohol intake converted to ethanol value

cancer, lung cancer, colon cancer, hepatocellular carcinoma,
and prostate cancer).
For HNC treatment, radiotherapy fields included the
lesion; the daily fractional radiation dose was 2 gray (Gy),
administered 5 days per week, for all patients. The standard
total radiation dose for HNC was 70 Gy. Chemotherapy for
HNC consisted of cisplatin only, 5-fluorouracil plus cispl-
atin, or S-1 only. One hundred and four patients (46.0%)
had undergone chemoradiation therapy as initial treatment
for HNC.
Comparison with characteristics of HNC patients
with or without ESCN
treatment for HNC was significantly lower in patients with
ESCN (14.7% vs. 35.0%, p = 0.030).
Characteristics of ESCN lesions
The characteristics of ESCN lesions are shown in Table 2.
Thirty-eight ESCNs were detected in 34 patients: 30 patients
had 1 lesion, and 4 patients had 2 lesions each. Thirty-three
synchronous and 5 metachronous ESCNs were diagnosed in
this study. Of the 38 ESCN lesions, 31 (81.6%) were early
ESCN (stage 0, I). These results demonstrated that ESCNs
were likely to be an earlier-stage neoplasia than HNC.
Treatment characteristics for patients with ESCN

There were 192 patients with HNC without ESCN (85.0%)
and 34 patients with HNC and with ESCN (15.0%) (Table 1).
Comparing the characteristics of HNC patients, the amount
of alcohol intake (44 vs. 20 g, p = 0.007) and location of
cancer in the hypopharynx (55.9% vs. 28.6%, p = 0.003)
were significantly greater in patients with ESCN compared
to those without ESCN. On the other hand, use of surgical
Treatments for HNC patients with ESCN are shown in
Table 3. Of 34 patients with ESCN, 10 patients (29.4%)
underwent endoscopic resection for ESCN and 10 patients
(29.4%) underwent simultaneous chemoradiation therapy.
Radiotherapy fields for ESCN included the lesion. The daily
fractional radiation dose was 2 Gy, administered 5 days per
week to all patients. The standard total radiation dose for

13
 Int J Clin Oncol

Table 2  Characteristics of esophageal squamous cell neoplasia The median follow-up period was 29.8 months (range,
(ESCN) 0–98.0 months). During the follow-up period, 17 (50.0%) of
No. of ESCN (%) HNC patients with ESCN and 57 (29.7%) of HNC patients
without ESCN died. Figure 1 shows the overall survival
Single/double
curves of HNC patients without or with ESCN. The median
 Single 30 (78.9)
survival period for HNC patients without ESCN was not
 Double 8 (21.1)
reached; that for those with ESCN was 45 months. The
Synchronous/metachronous
3-year survival rates for HNC patients with or without
 Synchronous 33 (86.8)
ESCN were 53% and 70%, respectively. Kaplan–Meier curve
 Methacronous 5 (13.2)
analysis using a log-rank test showed that the presence of
Clinical stage of ESCN (UICC 6th)
ESCN (p = 0.029) was associated with a greater probability
 Stage 0 11 (29.0)
of a shorter overall survival period.
 Stage I 20 (52.6)
We conducted univariate and multivariate analyses using
 Stage II 5 (13.2)
a Cox regression analysis with a time-dependent covariate
 Stage III 1 (2.6)
(Table 4). This analysis also identified a more advanced
 Stage IV 1 (2.6)
clinical stage (HR = 2.15; 95% CI = 1.18–3.93; p = 0.012)
ESCN esophageal squamous cell neoplasia, UICC Union for Interna- and the presence of ESCN (HR = 1.73; 95% CI = 1.00–2.97;
tional Cancer Control p = 0.049) as significant and independent determinants of
the overall survival period.

ESCN was 60 Gy. Chemotherapy for ESCN consisted of

5-fluorouracil plus cisplatin. Discussion

The high incidence of ESCN in HCN patients is illustrated

Impact of ESCN on survival outcome
To evaluate the actual magnitude of the impact of ESCN on
survival, the overall survival of HNC patients with or with-
out a prognostic factor [age: <68 years vs. ≥68 years; sex:
female vs. male; Brinkman index: <1000 vs. ≥1000; amount
of daily alcohol intake: <33 vs. ≥33 g; clinical stage: I, II
vs. III, VI; hypopharyngeal cancer: absence vs. presence;
ESCN: absence vs. presence; primary second malignancies
except ESCN: absence vs. presence; initial treatment of
HNC: local therapy (surgery, radiotherapy, other) vs. sys-
temic therapy (chemotherapy and chemoradiation therapy);
and chemoradiation therapy: absence vs. presence] was
compared.
by the concept of the “field cancerization” rule. Carcinogens
such as tobacco and alcohol elevate the risk of epithelial
cancer developing in the upper aerodigestive tract. In our
study, however, the incidence of ESCN in HCN patients was
at a higher rate than those of previous reports [6, 7]. Reasons
include the undertaking of routine EGDS in HNC patients as
well as endoscopic advances that are increasing the detec-
tion rates. In particular, endoscopic advances, such as high-
resolution endoscopy with magnification and narrow-band
imaging system, may contribute to the earlier detection of
lesions [16, 17].
In our study, the 3-year survival rate of HNC patients
with ESCN was more than 50%, which was an improvement
compared to that in past reports [12–14]. The contributing

Table 3  Types of treatments Treatment Head and neck cancer n (%) Esophageal

for tumors in patients with neoplasia n
esophageal squamous cell (%)
neoplasia (n = 34)
Surgical resection 3 (8.8) 1 (2.9)
Radiotherapy 3 (8.8) 0 (0.0)
Chemoradiation therapy 21 (61.8) 12 (35.3)
Chemotherapy and surgical resection 2 (5.9) 2 (5.9)
Radiotherapy and surgical resection 1 (2.9) 0 (0.0)
Chemoradiation therapy and surgical resection 2 (5.9) 0 (0.0)
Endoscopic resection 0 (0.0) 10 (29.4)
No treatment 2 (5.9) 9 (26.5)
Simultaneous chemoradiation therapy 10 (29.4)

13
Int J Clin Oncol

in HCN patients is critical to improving the prognosis [18];

Fig. 1  Kaplan-Meier survival curves comparing head and neck can-
cer patients (HNC) with or without esophageal squamous cell neo-
plasia (ESCN). p = 0.029 for a comparison between the groups by a
log-rank test
reasons include not only advances in anticancer therapies,
such as new surgical and radiotherapy techniques as well as
new anticancer drugs, but also the improved ability to detect
earlier ESCN lesions. In fact, the early detection of ESCN
most ESCNs in our study were detected at an early clinical
stage.
The development of ESCN, however, remained one of the
poor prognostic factors found for the overall survival rate
among the HNC patients in our study. The reasons for this
remain undetermined, but our data suggest several possibili-
ties. First, the rate of hypopharyngeal cancer patients with
ESCN was significantly higher than that of hypopharyngeal
cancer patients without ESCN. In our study, hypopharyn-
geal cancer was not a significantly poorer prognostic fac-
tor associated with overall survival of HNC patients. The
prognosis for patients with hypopharyngeal cancer, however,
has been relatively poor among patients with head and neck
cancers [19]. Therefore, it may affect the overall survival
of all patients with ESCN. Second, another possible expla-
nation for the shorter survival period in ESCN patients is
the adverse effect of simultaneous chemoradiation therapy
(CRT). The efficacy and feasibility of simultaneous CRT for
patients with HNC and ESCN have been reported [20–23].
Shinoto et al. showed that median survival was 19 months
and that the 2-year survival rate was 44% in patients with
synchronous HNC and esophageal tumors who received
simultaneous CRT; therapy was found to be feasible and
effective [23]. Simultaneous CRT, however, increased the

Table 4  Cox regression Risk factors Levels n Univariate analysis Multivariate analysis

analyses for risk factors in
overall survival rates HR 95% CI p HR 95% CI p

Age (years) <68 103 1

≥68 123 1.31 0.82–2.08 0.26
Sex Female 32 1
Male 194 0.77 0.37–1.61 0.49
Brinkman index <1000 160 1
>1000 66 0.67 0.39–1.16 0.15
Daily alcohol intake <33 g 130 1
>33 g 96 1.08 0.68–1.71 0.75
Clinical stage I, II 67 1 1
III, IV 159 2.21 1.21–4.03 0.010 2.15 1.18–3.93 0.012
Hypopharyngeal cancer Absence 152 1
Presence 74 1.09 0.66–1.76 0.75
ESCN Absence 192 1 1
Presence 34 1.81 1.05–3.11 0.033 1.73 1.00–2.97 0.049
Second primary tumor Absence 203 1
except ESCN Presence 23 1.18 0.59–2.36 0.65
Initial ­treatmenta Local treatment 104 1
Systemic treatment 122 1.51 0.94–2.42 0.085
Chemoradiation therapy Absence 122 1
Presence 104 1.39 0.88–2.20 0.15

HR hazard ratio, CI confidence interval, ESCN esophageal squamous cell neoplasia

a
  Local treatment; surgery, radiotherapy, and other; systemic treatment: chemotherapy and chemoradiation
therapy

13

radiation field of the lung and the adverse effects, resulting
in a reduced survival rate [24, 25]. In our series, the major-
ity of HNC patients with synchronous ESCN underwent
simultaneous chemoradiation therapy. Of the HNC patients
with synchronous ESCN, a definite CRT-related death was
observed in a patient with radiation pneumonitis; moreover,
the late complications of CRT were hidden as HNC-related
deaths. In fact, ESCN patients who received simultaneous
CRT tended to show a shorter overall survival period after
analysis using a log-rank test in our study (p = 0.10; data
not shown).
The present study had several limitations. First, patients
were enrolled in a single center. The findings reported here
should be confirmed in a multicenter study. A second limita-
tion was that the study design was retrospective: we relied
on medical records for patient data. Furthermore, the choice
of treatment regimen depended on the managing physicians.
The foregoing factors would have contributed to biases in
the present study.
In conclusion, ESCN is associated with a high incidence
of HNC patients so that screening for EGDS is crucial for
such patients. The presence of ESCN is one of the poor
prognostic factors in HNC patients, possibly because of the
presence of comorbid hypopharyngeal cancer and undergo-
ing simultaneous CRT.
Compliance with ethical standards  42:613–619
Conflict of interest  The authors declare that they have no conflict
of interest.
References
1. Ferlay J, Soerjomataram I, Dikshit R et al (2015) Cancer incidence
and mortality worldwide: sources, methods and major patterns in
GLOBOCAN 2012. Int J Cancer 136:E359–E386
2. Chuang SC, Scelo G, Tonita JM et al (2008) Risk of second pri-
mary cancer among patients with head and neck cancers: a pooled
analysis of 13 cancer registries. Int J Cancer 123:2390–2396
3. Schwartz LH, Ozsahin M, Zhang GN et al (1994) Synchronous
and metachronous head and neck carcinomas. Cancer (Phila)
74:1933–1938
4. Bongers V, Braakhuis BJ, Tobi H et  al (1996) The relation
between cancer incidence among relatives and the occurrence
of multiple primary carcinomas following head and neck cancer.
Cancer Epidemiol Biomark Prev 5:595–598
5. Johnson N (2001) Tobacco use and oral cancer: a global perspec-
tive. J Dent Educ 65:328–339
6. Slaughter DP, Southwick HW, Smejkal W (1953) Field canceriza-
tion in oral stratified squamous epithelium; clinical implications
of multicentric origin. Cancer (Phila) 6:963–968
7. Haughey BH, Gates GA, Arfken CL et al (1992) Meta-analysis
of second malignant tumors in head and neck cancer: the case
13
Int J Clin Oncol
for an endoscopic screening protocol. Ann Otol Rhinol Laryngol
101:105–112
8. Muto M, Hironaka S, Nakane M et al (2002) Association of mul-
tiple Lugol-voiding lesions with synchronous and metachronous
esophageal squamous cell carcinoma in patients with head and
neck cancer. Gastrointest Endosc 56:517–521
9. Su YY, Chen WC, Chuang HC et al (2013) Effect of routine
esophageal screening in patients with head and neck cancer.
JAMA Otolaryngol Head Neck Surg 139:350–354
10. Wang WL, Lee CT, Lee YC et al (2011) Risk factors for develop-
ing synchronous esophageal neoplasia in patients with head and
neck cancer. Head Neck 33:77–81
11. Hung SH, Tsai MC, Liu TC et al (2013) Routine endoscopy for
esophageal cancer is suggestive for patients with oral, oropharyn-
geal and hypopharyngeal cancer. PLoS One 8:e72097
12. Chen MC, Huang WC, Chan CH et al (2010) Impact of second
primary esophageal or lung cancer on survival of patients with
head and neck cancer. Oral Oncol 46:249–254
13. León X, Quer M, Diez S et al (1999) Second neoplasm in patients
with head and neck cancer. Head Neck 21:204–210
14. Chu PY, Chang SY, Huang JL et al (2010) Different patterns of
second primary malignancy in patients with squamous cell carci-
noma of larynx and hypopharynx. Am J Otolaryngol 31:168–174
15. Sobin LH, Wittekind C (2002) TNM classification of malignant
tumours, 6th edn. Wiley, Hoboken
16. Uedo N, Fujishiro M, Goda K et al (2011) Role of narrow band
imaging for diagnosis of early-stage esophagogastric cancer: cur-
rent consensus of experienced endoscopists in Asia-Pacific region.
Dig Endosc 23:58–71
17. Lee CT, Chang CY, Lee YC et al (2010) Narrow-band imaging
with magnifying endoscopy for the screening of esophageal can-
cer in patients with primary head and neck cancers. Endoscopy
18. Lim H, Kim DH, Jung HY et al (2015) Clinical significance of
early detection of esophageal cancer in patients with head and
neck cancer. Gut Liver 9:159–165
19. Cooper JS, Porter K, Mallin K et al (2009) National Cancer Data-
base report on cancer of the head and neck: 10-year update. Head
Neck 31:748–758
20. Guillot T, Spielmann M, Kac J et al (1992) Neoadjuvant chemo-
therapy in multiple synchronous head and neck and esophagus
squamous cell carcinomas. Laryngoscope 102:311–319
21. Nguyen TD, Panis X, Legros M et al (1989) Neoadjuvant chemo-
therapy and irradiation in multiple synchronous squamous cell
carcinoma of the upper aerodigestive tract. Radiother Oncol
16:283–288
22. Welz S, Schmid A, Hehr T et al (2005) Treatment-outcome for
synchronous head-and-neck and oesophageal squamous cell car-
cinoma. Radiother Oncol 77:267–270
23. Shinoto M, Shioyama Y et al (2011) Clinical results of definitive
chemoradiotherapy for patients with synchronous head and neck
squamous cell carcinoma and esophageal cancer. Am J Clin Oncol
34:362–366
24. Schallenkamp JM, Miller RC, Brinkmann DH et al (2007) Inci-
dence of radiation pneumonitis after thoracic irradiation: dose–
volume correlates. Int J Radiat Oncol Biol Phys 67:410–416
25. Vogelius IR, Bentzen SM (2012) A literature-based meta-analysis
of clinical risk factors for development of radiation induced pneu-
monitis. Acta Oncol 51:975–983