вход по аккаунту



код для вставкиСкачать
Correlation between Computed Tomographic Density
of Lymph Node Metastases and Response to
Cisplatin-Based Chemotherapy in Patients with Head
and Neck Squamous Cell Carcinoma in an Area in
which Betel Quid Chewing Is Prevalent
Hung-Ming Wang, M.D.’
Shu-Hang Ng, M . D . ~
Cheng-Hsu Wang, M.D.’
Chuang-Chi Liaw, M.D.’
Ming-Hsui Tsai, M . D . ~
Gi-Ming Lai, M . D . ~
’ Division of Hematology/Oncoiogy, Department
of Internal Medicine, Chang Gung Memorial
Hospital, Lin-Kou Medical Center, Taipei, Taiwan, Republic of China.
* Department of
Diagnostic Radiology, Chang
Gung Memorial Hospital, Lin Kou Medical Center, Taipei, Taiwan, Republic of China.
Department of Head and Neck Surgery, Chang
Gung Memorial Hospital, Lin-Kou Medical Center, Taipei, Taiwan, Republic of China.
The preliminary result (with fewer patients) of
this study was presented at the Annual Meeting
of The Chinese Oncology Society, Taipei, Taiwan, Republic of China, April 16, 1995.
Address for reprints: Hung-Ming Wang M.D.,
Division of Hematology/Oncology, Department
of Internal Medicine, Chang Gung Memorial
Hospital, 199, Tun Hwa North Road, Taipei 105,
Taiwan, Republic of China.
Received February 7, 1996; revisions received
May 21, 1996, and July 5, 1996; accepted July
5, 1996.
0 1996 American Cancer Society
BACKGROUND. Identifying the factors predicting response to chemotherapy is imporlant for patients with head and neck squamous cell carcinoma (HNSCC). It
allows more rational selection of subsets of patients who may benefit from multidisciplinary treatment. Correlation of lymph node density in contrast-enhanced
computed tomographic (CT) scans of HNSCC with response to Chemotherapy was
observed in the recent literature. This prospective study was designed to validate
this clinical issue.
METHODS. From January 1992 to March 1995, 71 patients with untreated HNSCC
were included in this study in which the following criteria were met: 1) a lymph
node > 3 cm by physical examination or z 2 cm by scanographic examination;
2) clinically evaluable disease treated by cisplatin-based neoadjuvant chemotherapy; and 3) availability of a pretherapeutic contrast-enhanced CT scan showing
the cross-sections of relevant lymph node metastases. The density of the largest
lymph node was compared with that of the nuchal muscles by a radiologist blinded
to the patient’s therapeutic outcome. A lymph node was classified as hypodense
if more than 33% of the lymph node surface area was comprised of a hypodense
zone, and isodense if less than a third of the lymph node surface area was comprised of a hypodense zone.
RESULTS. Fifty-one patients (72%) had the largest lymph node classified as hypodense, and 63 patients (89%) were found to have extranodal spread (ENS) in the
relevant lymph nodes. Fifty-nine patients were betel quid chewers. Lymph node
density was not related to T classification, primary site, or histologic differentiation
of the primary tumor. There was also no correlation between lymph node density
and lymph node N classification. The lymph node chemotherapy response rate
was 35% (7 of 20) of the isodense group and 47.1% (24 of 51) of the hypodense
group ( P = 0.36).ENS was found to have no impact on the chemotherapy response.
CONCLUSIONS. Computed tomographic density of lymph node metastases did predict chemotherapy response in the HNSCC patients in the current study from an
area in which betel quid chewing is prevalent. Cancer 1996; 78:1972-9.
0 1996 American Cancer Society.
KEYWORDS: head and neck neoplasms, cervical lymph node, chemotherapy, betel
ead and neck squamous cell carcinoma (HNSCC) (excludingnasopharyngeal carcinoma) accounts for 4-5% of the cancer incidence
in Taiwan.’ The majority of the patients with HNSCC present with
Lymph Node Density and Drug Response in HNSCCMlang et al.
Stage 111 or IV disease. Despite optimal local therapy,
long term survival rates for these patients remain less
than 30%. During the last 10 years, chemotherapy has
been actively used in conjunction with surgery and/
or radiotherapy in the management of patients with
advanced stage head and neck carcinoma. Nonrandomized trials applying neoadjuvant chemotherapy in
advanced stage HNSCC patients have demonstrated
encouraging results with 60-90% tumor regression
and 20-50% complete tumor regre~sion.~-~
none of the randomized trials using neoadjuvant chemotherapy has demonstrated increased survival
rate^.^-^ Neoadjuvant chemotherapy would only benefit the responders, in terms of enhancement of locoregional control, and preservation of anatomic function
in conjunction with c o s m e s i ~ . ~Therefore,
f ~ ' ~ f ~ to identify the predicting factors for response to chemotherapy would be helpful in selecting suitable patients to
receive neoadjuvant chemotherapy, which is considered to be relatively time-consuming and potentially
toxic multidisciplinary management.
The reported prognostic indicators for response
to neoadjuvant chemotherapy included performance
status (PS), primary tumor site, extent of lymph node
involvement, and tumor size.l0-" The authors' previous study revealed that the tumor size and the presence of lymph node metastases were the most important predictors for response to neoadjuvant chem ~ t h e r a p y .Predictive
factors for chemosensitivity in
patients with lymph node metastases are important
for medical oncologists to justify the multimodality
therapeutic sequence.
A relationship between responses to chemotherapy and the density of lymph node metastases in contrast computed tomographic (CT) scans was demonstrated in recent trial^.'^.'^ Complete response rates
from chemotherapy were worse if greater than 33% of
the metastatic lymph node area was occupied by a
hypodense zone. This phenomenon was attributed to
the functional vascularization of the lymph nodes for
efficient drug delivery.
Epidemiologic studies have indicated that betel
quid chewing appears to be an important factor in the
development of HNSCC, especially oral cavity carcinoma, in Taiwan.l6-lEAccording to a recent survey,
approximately 2.8 million people (approximately
13.3%of the Taiwanese population) are currently betel
quid chewers. In Taiwan, betel quid chewers carry a
relative risk 28 times higher than average for oral carcinoma development when compared with individuals
who do not chew betel quid. If the betel quid users
also consume cigarettes and alcohol, the relative risk
increases 123 times. Approximately 86% of the Taiwanese betel quid users were also cigarette smokers
and 75% also consumed alcohol habit~ally.'~
The responses to neoadjuvant chemotherapy of the authors'
HNSCC patients (80-90% of whom were betel quid
chewers) were not as good as those reported from the
Western ~0untries.l~
The major characteristics of these
patients that differed from those in the literature included the predominance of oral cavity carcinoma, a
high incidence of severe mucositis, and bulky tumor
burden. These characteristics were closely related to
the culture of betel quid chewing, and may account
for the lower response rates to chemotherapy of the
authors' HNSCC patients. Therefore, this prospective
study was designed to evaluate whether the correlation
between the lymph node density in contrast CT scans
and response to chemotherapy could be demonstrated
in HNSCC patients from a betel quid chewing area.
The histologic demonstration of extranodal spread
(ENS) of the tumor in cervical metastasis has, according to many authors, a negative impact on the
outcomes of HNSCC.20-2'However, nothing in the literature mentioned the relationship between ENS and
chemotherapy response. In this article, ENS was identified on contrast CT scan and the possible role of
ENS in the lymph node response to chemotherapy was
From January 1992 to April 1995, 230 patients with
HNSCC were included in prospective neoadjuvant
chemotherapy trials. Patients eligible for these trials
were required to have histologically confirmed squamous cell carcinoma; a measurable or evaluable locoregional disease with no distant metastasis: and no
prior treatment. All patients had a World Health Organization (WHO) PS of 0-2, adequate renal function
(serum creatinine < 1.5 mg/dL), leukocyte counts =4000/mL, and platelet counts 2 100,00O/mL.The pretreatment staging procedures were comprised of history and physical examination, complete blood count
and biochemistry, chest roentgenography, bone scan,
ultrasonography of liver, and CT scan of the head and
neck. The chemotherapy regimens included 1) cisplatin (CDDP), 100 mg/m2 per day on Day 1; and 5fluorouracil (5-FU), 1 gm/m2 continuous intravenous
infusion, on Day 1 to Day 4 or 5; or 2) CDDP, 100 mg/
m2, on Day 1: 5-FU, 1 gmlm' per day continuously
intravenous infusion, on Day 1 to Day 4; and methotrexate (MTX), 200 mglm', on Days 15 and 22 with
oral calcium folinate, 15 mg every 6 hours, rescue for
3 days (starting at 24 hours after MTX). Both regimens
were given every 4 weeks.
Patients who met the following criteria proposed
by Janot et a1.15 were included in this study: clinically
measurable lymph node > 3 cm or lymph node scano-
CANCER November 1, 1996 / Volume 78 / Number 9
FIGURE 1. Axial postcontrast computed tomography scan shows an
extremely large isodense lymph node involved the left upper jugular area
with extranodal spread. The common carotid artery (large arrow) is encased by the tumor lymph node and the jugular vein is not visible. Foci
of hypodense lymph nodes (arrows) are seen within the aggregation of
lymph nodes.
graphic greatest dimension 2 2 cm; clinically evaluable disease treated by CDDP-based neoadjuvant chemotherapy; and availability of a pretherapeutic contrast CT scan obtained within 2 weeks before the
treatment initiation and showing cross-sections of relevant metastatic lymph nodes.
The CT scans were performed with GE CT9800 or
GE CT Prospeed Plus machines (GE Medical System,
Milwaukee, WI). Iodine-based contrast material (Hypaque-76, a brand of diatrizoate meglumine and diatrizoate sodium, Sanofi Winthrop Pharmaceuticals,
NY) was injected by intravenous bolus at a dosage of
100 mL. The primary tumor and the clinically relevant
lymph nodes were scanned in the axial plane with 5mm thickness. The pretherapeutic contrast CT scans
for staging purpose were analyzed by the radiologist
in charge of the head and neck imaging section at
the current study hospital, who was blinded to the
treatment choice and therapeutic outcomes. The
largest visible lymph node greater than 2 cm in greatest dimension was chosen for the grading purpose.
The density of normal adjacent muscle tissue was chosen as the density reference. Lymph node density was
graded according to the classification proposed by
Munck et al.14A lymph node was classified as isodense
if less than one-third of the lymph node was com-
FIGURE 2. Axial postcontrast computed tomography scan shows an
extremely large hypodense lymph node with extranodal spread on the right
side. middle neck.
prised of hypodense zones (Fig. 1) and as hypodense
if one-third or more of the lymph node was comprised
hypodense zones (Fig. 2). The lymph nodes with peripheral capsular enhancement and relative central
hypodensity were graded as isodense if the central
lymph node architecture was not more hypodense
than the adjacent muscles (Fig. 3). When multiple
lymph nodes were present and of a similar size with
different density characteristics, the overall grading
was considered as hypodense.
The ENS was identified on the contrast CT scans
by the radiologists as indistinct margins, thickened adjacent muscle or deep fasciae, an enhancing lymph
node rim, or infiltration of the adjacent fat planes.23
Response was evaluated by clinical and radiologic
studies after two cycles of chemotherapy. Nonresponders and those with progressive disease (PD) immediately underwent locoregional treatment. The responders were given a third cycle of chemotherapy
and were then reevaluated for response before receiving definite locoregional treatment. Definition of response was defined by the WHO
A complete
response (CR) was defined if there was no evidence of
tumor on gross palpation andlor by CT scan. A partial
response was defined as 250% decrease in the product
of the longest tumor dimension and its greatest perpendicular diameter and no increase in the size of any
other known disease. Tumor regression < 50% of the
Lymph Node Density and Drug Response in HNSCC/Wang et al.
TNM Classification"of the Patients
'American Joint Committee on Cancer, 198R.
Lymph Node Density and Tumor Characteristics
P value
1.15 I2.27
3.87 I1.70
FIGURE 3. Axial postcontrast computed tomography scan shows an
extensively enlarged lower jugular lymph node with peripheral capsular
enhancement and relative central hypodensity. It was graded as isodense
because the central lyniph node parenchyma was not more hypodense
than the adjacent muscles. It also demonstrated extranodal spread.
initial tumor size was termed stable disease. PD was
the product of the two greatest dimensions that
showed 225% increase over the initial product, or evolution of a new lesion.
Factors evaluated for their correlation with lymph
node density included tumor site, histologic differentiation, 1' and N categories, and ENS. The T and N categories, lymph node density, and ENS were evaluated
for their correlation with chemotherapy response. The
chi-square tests and Fisher's exact test were used to
analyze statisticall significance of qualitative parameters and the Student's t test of quantitative parameters.
The level is accepted as significant if the P value <
Seventy-one patients met the eligible criteria and were
included in this study. The median age of the patients
was 50 years (range, 30-74 years) and the sex ratio
(maleifemale) was 70: 1. Approximately 86%were betel
quid chewers, 90% were cigarette smokers, and 93%
consumed alcohol. The primary tumor sites were the
following: 9 in the oropharynx, 36 in the oral cavity,
25 in the hypophaxynx, and 1 in the larynx. TNM distribution is shown in Table 1 (according to the American
Joint Committee for Cancer Staging 1988"). The neoadjuvant chemotherapy regimen comprised of CCDP
+ 5-FU was used in 44 patients, and CDDP + 5-FU +
MI'X in 27 patients.
Primary site
Oral cavity
Histologic differentiation
Primary tumor extent
Lyniph node size
Extranodal spread
Mean size of lvmuh node icml
Fifty-one patients (72%) were classified as having
hypodense lymph nodes and the remaining 20 patients as having isodense lymph nodes. The possible
correlation between lymph node density and other
factors such as primary site, histologic differentiation,
T and N categories, and ENS was analyzed and is
shown in Table 2. Eleven patients were referred from
other hospitals without knowledge of the tumor differentiation. There was no statistically significant correlation between hypodensity grade and primary tumor
site, histologic differentiation, or mean size of lymph
nodes. The incidence of ENS was 89%, which was
lower in the isodense group than in the hypodense
group. The mean lymph node size was 4.16 2 1.87 cm
and 2.25 2 0.35 cm in the lymph nodes with and without ENS, respectively ( P < 0.001).
Correlation analysis of chemotherapy response
with other factors was shown in Table 3 . There were
no significant differences between T and N classification in correlation with chemotherapy response in
CANCER November 1, 1996 / Volume 78 / Number 9
Chemotherapy Response and Tumor Characteristics
Comparative Differences Between Studies
Patient no.
Responders ( W )
P value
16 (76.2%)
29 (58%)
21 (47.7%)
10 (37%)
7 (35%)
24 (47.1%)
26 (41.2%)
5 162.5%1
Janot et al.
W a g et al.
No. of patients
Tumor site
Oral cavity
T stage
T1-2:T3-4 (%)
N stage
Primary tumor
TI -3
Lymph node
Lymph density
Extranodal spread
these patients with lymph node metastases. The response rate of the isodense lymph nodes (35%) was
not significantly different from that of the hypodense
lymph nodes (47.1%)( P = 0.36).Also, no difference in
the response rate was observed between the lymph
nodes with or without ENS.
Hypodense lymph node
% of all patients
% of N1-2:% of N3
Mean lymph node size (cm)
lsodense lymph node
Hypodense lymph node
Chemotherapy response
Isodense lymph node (%)
Hypodense lymph node (W)
% Of N1-2:% of N3
3.3 ? 0.6
2.5 ? 0.5
4.2 ? 2.3
3.9 ? 1.7
Six patients had second primaries, total 42 primary sites.
Neoadjuvant chemotherapy has been widely accepted
as an important part of the multidisciplinary management for patients with advanced stage HNSCC, particularly for the inoperable patients or for the purpose
of organ p r e ~ e r v a t i o n . ~Because
~ ~ ’ ~ - ~only responders
may benefit from this time-consuming and potentially
toxic therapeutic sequence, identifymg any predictive
factors of response to chemotherapy before treatment
is important for more rational patient selection.
In Taiwan, 80-90% of HNSCC patients chew betel
quid habitually. The authors’ previous neoadjuvant
chemotherapy trials revealed some characteristics that
were different from the literature reports of the Western society. Their findings revealed that the primary
tumor site was predominantly in the oral cavily, especially the buccal mucosa and the hard palate; there
was a high prevalence of WHO Grade 3-4 chemotherapy-related mucositis (40% of patients) and more
bulky tumor burden.’lz6The response rate to the CDDP
and 5-FU combination regimen in the 96 patients was
56% with only 3% CR.13 The low response rate may
be due to the inadequate dose intensity resulting from
betel quid chewing-related mucosa damage and mucositis, and bulky tumor secondary to delayed management because of the low socioeconomic level of
betel quid c h e ~ e r s . ’ ~ ~ ~ ~ ~ ~ ~
Iodinated contrasted CT scan demonstrated semiquantitative reflection of blood perfusion; therefore,
the lymph node density is likely to represent distribu-
tion of the contrast agent in the vasculature of the
lymph node. Because an efficient vascular flow is necessary for homogeneous drug delivery in tumors, it
appears reasonable to link chemotherapy response
with lymph node density.I4,l5The patients with regional lymph node metastases demonstrated even
worse chemotherapy response in the current study
and other literature^.'^^'^^^^ It was interesting to learn
whether lymph node density could be a predictive factor for chemotherapy response of metastatic lymph
nodes in the study patients.
Only the lymph nodes with a scanographic greatest dimension 2 2 cm were investigated, because the
false-positive rate of metastatic lymph nodes at this
The same grading
size is considered low (4%-7%).31,32
method of lymph node density was used as that according to the report of Munck et al.14for the purpose
of comparison. However, the current study failed to
confirm the correlation between lymph node density
and chemotherapy response.
The patient characteristics from the current
study were compared with those from Janot et al.15
and are presented in Table 4, both studies focused
on HNSCC patients only. The current series had
much more oral cavity HNSCC patients (51%) than
those from the studies by Munck et al.’* (12%) and
(21 of 71) of the patients in
Janot et al.I5 ( l l % )30%
lymph Node Density and Drug Response in HNSCClWang et al.
the current study had as primary sites the buccal
mucosa and hard palate.
In addition to the availability of drug delivery to
the target, another factor affecting the capacity of a
drug to kill the cancer cell is the cell's biochemical
resistance to the
The probability of a given
tumor containing resistant clones, when a patient is
newly diagnosed, would be a function of the mutation
rate and the size of the tumor.34The incidences of
lymph node metastasis at presentation were lower in
patients with the buccal mucosa as the primary site
of HNSCC (9-31%) and those whose primary site was
the hard palate (13-24%) than those whose primary
site was the pharynx (50-80%).35 Most of the oral cavity HNSCC patients enrolled in this study with lymph
nodes 2 2 cm in greatest dimension had bulky tumor
burden in their primary sites. Therefore, there may be
more chemoresistant clones in oral cavity HNSCC and
its metastatic lymph nodes. In addition, as previously
described, the impact from betel quid chewing further
negatively influenced lymph node chemotherapy response. These lower lymph node response rates to
chemotherapy may overshadow the significant prognostic indicator of CT lymph node density.
The other factor that may account for the different
results may be the intrinsic defect of the grading system used, which was proposed by Munck et al.14Why
was one-third of the lymph node's density considered
the critical factor under this grading system, rather
than one-fourth or one-fifth? A necrotic lymph node
was a necrotic lymph node, and to classify them into
different groups (isodense or hypodense) according to
the density variations might be a false indicator and
demonstrate biased data. In addition, using the largest
visible lymph node greater than 2 cm in dimension
for grading purposes may not correlate well with the
outcome. Many HNSCC patients had multiple lymph
nodes with various densities (Fig. l), and their outcomes might be affected largely by the smaller necrotic
lymph nodes with poor response to treatment modalities rather than the large isodense lymph nodes. Another potential error was to grade a necrotic lymph
node as isodense if the density of the central lymph
node parenchyma was equivalent to adjacent muscles
(Fig. 3). It was not only difficult to classify a necrotic
lymph node into the isodense group, but in this situation it also might be related to the technique of CT
scan. This type of grading system biased the data and
demonstrated why the response of the patients in the
current study with isodense lymph nodes was not better than those with hypodense lymph nodes.
In addition to the issues mentioned above, the
grading method used in these trials cannot reflect the
truly viable tumor burden, which might be closely re-
lated to the chemotherapy response within different
lymph nodes.33The central lucent area of the CT scan
in hypodense lymph nodes usually contains both tumor infiltration and necrosis. Whereas both of these
pathologic changes cause decreased lymph attenuation at CT scan, they cause different signal intensities
in magnetic resonance imaging. Most tumor cells have
intermediate signal intensity on T1- and T2-weighted
images. Necrotic areas have low T1-weighted and very
high T2-weighted signal i n t e n ~ i t y . ~These
~ . ~ ' changes
are admixed with the signal intensity of the uninvolved
lymph node tissue, leading to a complex interplay of
signal intensities. Because the CT scan cannot truly
demonstrate the relative amount of lymph node necrosis that varies from lymph node to lymph node with
a hypodense center, it is not surprising that the lymph
node density of contrast CT scan correlated poorly
with chemotherapy response. Also, the isodense
lymph nodes may contain more viable tumor cells
than hypodense lymph nodes with similar scanographic greatest dimension. This contradiction between lymph node density and viable tumor mass implies the over simplification to correlate lymph node
density with chemotherapy response.
Both the trial by Munck et al.I4 and the current
study revealed that the hypodense aspect was independent of lymph node size, and also demonstrated
that greater than 70% of lymph node metastases were
graded as hypodense. However, the report by Janot et
al.,15the isodense lymph nodes were shown to have
a larger scanographic dimension, N3 disease had a
significantly higher isodense rate than N1-2 disease.
Consequently, the chemotherapy response rate of N3
disease was higher than that of N1-2 disease. This
result was different from those in the literature and
deserved to be ~ l a r i f i e d . " ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~
Although the histologic demonstration of ENS of
tumor in cervical metastasis was associated with a
poor outcome in patients with HNSCC, no correlation
between the ENS and the lymph node response to
chemotherapy was demonstrated in this trial. In
lymph nodes larger than 2 cm in greatest dimension,
ENS is present in greater than 50%.23The fact that 89%
of the patients in the current study had ENS implied
poor response to treatment modalities and poor outcomes. The great discrepancy between the numbers
of patients with or without ENS may preclude a meaningful statistical analysis.
In conclusion, the current study failed to validate that
lymph node density could be a predictor for chemotherapy response of metastatic lymph nodes in
HNSCC patients. The variable response rates of lymph
CANCER November 1, 1996 / Volume 78 / Number 9
nodes were affected by multiple factors. Betel quid
chewing behavior resulting in differences in primary
tumor sites, dose intensity, and tumor burden may
contribute to these conflicting results from different
studies. There were also many debatable i,;(sues regarding the grading system proposed by Munck et al.,I4
and these might affect the eventual outcomes. Future
trials with adequate patient population and a more
clear method €or lymph node density grading may be
needed before using lymph node density for patient
selection in multidisciplinary managemenit of advanced HNSCC.
Department of Health, the Executive Yuan, R O C . Cancer
Registry Annual Report in Taiwan Area, 1990. June 1994.
Dimery IW, Hong WK. Overview of combined modality therapies for head a n d neck cancer. J Natl Cancer Inst
Schantz SP, Harrison I,B, Hong WK. Cancer of head and
neck. In: Devita VT, IIellman S, Rosenberg SA, editors. Cancer: principles & practice of oncology. 4th edition. Philadelphia: J.B. Lippincott, 1993:574-630.
Toohill RJ, Anderson T, Byhardt RW, Cox JD, Duncavage JA,
Grossman TW, et al. Cisplatin and fluorouracil as neoadjuvant chemotherapy in head and neck cancer. Arch Otolaryngo/ Head Neck Surg 1987; 113:758-GI.
The Department of Veterans Affairs Laryngeal Cancer Study
Group. Induction chemotherapy plus radiation compared
with surgeiy plus radiation in patients with advanced laryngeal cancer. iVEngl J Men 1991;324:1685-90.
Paccagnella A, Orlando A, Marchiori C, Zorat PL, Cavaniglia
G, Sileni VC, et al. Phase 111 trial of initial chemotherapy in
stage 111 or IV head and neck cancers: a study by the Cruppo
di Studio sui Turnori della Testa e del Collo. / Natl Cancer
Martin M, Malaurie E, Langlet PM, Vergnes L, 1.elievre G,
Mazeron JJ, et al. A randomized prospective study of CDDP
and 5-FU as neoadjuvant chemotherapy in head a n d neck
cancer: a final report. Proc A m Soc Clin O n c d 1995; 14:294.
Jacobs C, Goffniet DR, Goffinet L, Kohler M, Fee WE. Chemotherapy as a substitute for surgery in the treatment of advanced resectable head and neck cancer. Cancer 1987;
Pfister DG, Strong E, Harrison I., Haines IE, Pfister DA, Sessions 11, et al. Larynx preservation with combined chemotherapy and radiation therapy in advanced but resectable
head and neck cancer. J Clin Oncol1991;9:850-9.
Cognetti F, Pinnaro P, Ruggeri EM, Carlini P, Perrino A, Impiombato FA, et al. Prognostic factors for chemotherapy response and survival using combination chemotherapy as
initial treatment of advanced head and neck squamous cell
cancer. J Clitz Oncol 1989;7:829-37.
Wolf G l , Makuch IIW, Raker SR. Predictive factors for tumor
response to preoperative chemotherapy in patients with
head and neck squainous carcinoma: the head and neck
contracts program. Cuncer 1984;54:2869-77.
Clavel M, Mansour AR. Head and neck cancer: prognostic
factors for response to chemotherapy. Eur I Cuncer
Wang HM, Wang CH, Chen JS, Chang HK, Kiu MC, h a w
CC, et al. Cisplatin and 5-FU as neoadjuvant chemotherapy:
predicting response in head anti neck squamous cell cancer.
/ Forrnos Med Assoc 1995;94:87-94.
Munck JK, Cvitkovic E, Piekarski ID, Benhamou E, Recondo
G, Bachouchi M, et al. Computed tomographic density of
metastatic lymph nodes as a treatment-related prognostic
factor in advanced head and neck cancer. / Nut1 Cancer lnst
Janot F, Cvitkovic E, Piekarski JD, Sigal R, Armand JP, Bensm a h e A, et al. Correlation between nodal density in contrasted scans an d response to cisplatin-based chemotherapy
in head and neck squamous cell cancer: a prospective validation. Head Neck 1993; 15222-9.
Chong KM. Betel nut chewing and mouth cancer in Taiwan.
Observation of the oral mucosa in the betel nut chewers. J
Forrnos Med Assoc 1966;65:79-85.
Kwan HW. A statistical study o n oral carcinomas in Taiwan
with emphasis o n the relationship with betel nut chewing:
a preliminary report. J Fornzos Med Assoc 1976;75:497505.
Kuo MW, Jeng JH, Chiang CP, Hahn LJ. Mutations of Ki-ras
oncogene codon. 12 in betel quid chewing-related human
oral squamous cell carcinoma in Taiwan. J Orul Patho/ Med
KO YC. A review of betel quid chewing and its health effects
in Taiwan. Symposium on betel quid chewing and its
health effect [in Chinese] [abstract]. Krrolisiiing R.O.C.
20. Johnson JT, Barnes L, Myers EN. The extracapsular spread
of tumors in cervical node metastasis. ;Irclz Otolnryngol
21. Snow GB, Annyas AA, VanSlooten AE, Rartelink H, Hart
AAM. Prognostic factors of neck node metastasis. Clin.O t o h
vyngol 1982;7:185-92.
22. Johnson JT, Myers EN, Bedetti CD, Barnes EL, Schrainm VL,
Thearle PB. Cervical lymph node metastases. Arch Otolar~mgo1 Head Neck Surg 1985; 11 1 :534-7.
23. Son1 PM. Lymph nodes. In: Som PM, Curtin HD, editors.
Head and neck imaging. 3rd edition. St. Louis: Mosby-Year
Book, Inc., 1996:772-93.
24. Miller AB, Hoogstraten B, Staquet M, Winkler A. Reporting
results of cancer treatment. Cancer 1981;47:207-14.
25. American Joint Committee on Cancer. Manual for sraging
of cancer. 3rd edition. Philadelphia: J.B. Lippincott, 1988.
26. Muir C, Weiland I.. Upper aerodigestive tract cancer. Cuncer
1995;75: 147-53.
27. Rooney M, Kish 1, Jacobs J, Kinzie J, Weaver A, Crissman
J, et al. Improved complete response rate and survival in
advanced head and neck cancer after three-course induction
therapy with 120-hour 5 - F U infusion and cisplatin. Cancer
28. Ko YC, Chiand TA, Chang SJ, Hsieh SF. Prevalence of betel
quid chewing habit in Taiwan and related sociodemographic
factors. J Oral Putliol Med 1992;21:261-4.
29. Bachaud JM, David J M , Shubinski RF, Perineau D, Boussin
G, Serrano E, et al. Predictive factors of a complete response to an d adverse effects of a CDDP-5-FU conibination as primary therapy for head an d neck squamous carcinomas. / I,aryngo/ O f o l 1993;107:924-30.
30. ‘l‘hyss A, Schneider M, Santini J , Caldani C, Vallicioni J,
Chauvel P, et al. Induction chemotherapy with cisplatin a nd
5-fluorouracil for squamous cell carcinoma of the head and
neck. Br J Cancer 1986;54:755-60.
Lymph Node Density and Drug Response in HNSCCMlang et al.
31. Moreau P, Goffard Y, Collignon J. Computed tomography of
metastatic lymph nodes: a clinical, computed tomographic,
pathologic correlative study. Arch Otolaryngol Head Neck
Surg 1990;116:1190-3.
32. Rainer T, Ofner G, Marckhgott E. Ultrasound diagnosis of
regional lymph node metastasis of the neck in patients with
head-neck neoplasm: sono-morphologic criteria and diagnostic accuracy. Laryngorhinootologie 1993;7273-7.
33. Devita VT. Principles of chemotherapy. In: Devita VT, Hellman S, Rosenberg SA, editors. Cancer: principles & practice
of oncology. 4th edition. Philadelphia: J.B. Lippincott,
34. Goldie JH, Coldman AJ. A mathematic model for relating
the drug sensitivity of tumors to the spontaneous mutation
rate. Cancer Treat Rep 1979;63:1727-33.
35. Million RR, Cassisi NJ, Mancuso AA, Stringer SP, Mendenhall
WM, Parsons JT. Management of the neck for squamous cell
carcinoma. In: Million RR, Cassisi NJ, editors. Management
of head and neck cancer: a multidisciplinary approach. 2nd
edition. Philadelphia: J.B. Lippincott, 1994:75- 142.
36. Mancuso AA, Dillon WP. The neck. Radio1 Clin North A m
37. Yousem DM, Sam PM, Hackney DB, Schwaibold F, Hendrix
RA. Central nodal necrosis and extracapsular neoplastic
spread in cervical lymph nodes: MR imaging versus CT. Radiology 1992;182:753-9.
38. Davis RK, Stoker K, Harker G, Davis K, Gibbs FA, Harnsberger HR, et al. Prognostic indicators in head and neck cancer patients receiving combined therapy. Arch Otolaryngol
Head Neck Surg 1989;115:1443-6.
Без категории
Размер файла
774 Кб
Пожаловаться на содержимое документа