11460.Oral Cancer Research Advances by Alexios P. Nikolakakos

8/17/2019 11460.Oral Cancer Research Advances by Alexios P. Nikolakakos

1/320


2/320

ORALCANCERRESEARCHADVANCES

No part of this digital document may be reproduced, stored in a retrieval system or transmitted in any form or

by any means. The publisher has taken reasonable care in the preparation of this digital document, but makes no

expressed or implied warranty of any kind and assumes no responsibility for any errors or omissions. No

liability is assumed for incidental or consequential damages in connection with or arising out of informationcontained herein. This digital document is sold with the clear understanding that the publisher is not engaged in

rendering legal, medical or any other professional services.


3/320


4/320

ORAL CANCER RESEARCH ADVANCES

ALEXIOS P. NIKOLAKAKOS

EDITOR

Nova Biomedical Books

New York


5/320


6/320

Copyright © 2007 by Nova Science Publishers, Inc.

All rights reserved. No part of this book may be reproduced, stored in a retrieval system or

transmitted in any form or by any means: electronic, electrostatic, magnetic, tape, mechanical

photocopying, recording or otherwise without the written permission of the Publisher.

For permission to use material from this book please contact us:

Telephone 631-231-7269; Fax 631-231-8175

Web Site: http://www.novapublishers.com

NOTICE TO THE READER

The Publisher has taken reasonable care in the preparation of this book, but makes no expressed

or implied warranty of any kind and assumes no responsibility for any errors or omissions. No

liability is assumed for incidental or consequential damages in connection with or arising out of

information contained in this book. The Publisher shall not be liable for any special,

consequential, or exemplary damages resulting, in whole or in part, from the readers’ use of, or

reliance upon, this material.

Independent verification should be sought for any data, advice or recommendations contained in

this book. In addition, no responsibility is assumed by the publisher for any injury and/or damage

to persons or property arising from any methods, products, instructions, ideas or otherwise

contained in this publication.

This publication is designed to provide accurate and authoritative information with regard to thesubject matter covered herein. It is sold with the clear understanding that the Publisher is not

engaged in rendering legal or any other professional services. If legal or any other expert

assistance is required, the services of a competent person should be sought. FROM A

DECLARATION OF PARTICIPANTS JOINTLY ADOPTED BY A COMMITTEE OF THE

AMERICAN BAR ASSOCIATION AND A COMMITTEE OF PUBLISHERS.

LIBRARY OF CONGRESS CATALOGING-IN-PUBLICATION DATA

Oral cancer research advances / Alexios P. Nikolakakos, editor.

p. ; cm

Includes bibliographical references and index.

ISBN 13: 978-1-60741-924-2 (E-Book)

1. Mouth--Cancer. 2. Head--Cancer. 3. Neck--Cancer. I Nikoladados, Alexios P.

[DNLM: 1. Mouth Neoplasms. 2. Head and Neck Neoplasms. 3. Mouth Neoplasms--

genetics. WU 280 0632 2007]

RC280. M60725 2007-11-07

616.99’491--dc22

2007026603

Published by Nova Science Publishers, Inc. New York


7/320

CONTENTS

Preface ix

Chapter 1 Effective Administration Methods of 5-Aminolevulinic Acid as a

Photosensitizer in Photodynamic Therapy for Tongue Tumor 1 Toshiyuki Ogasawara, Norio Miyoshi, Kazuo Sano, Hidetaka Kinoshita,

Tetsushi Yamada, Toru Ogawa, Kazuki Miyauchi and Yoshimasa Kitagawa

Chapter 2 Relationships Between Biological and Clinicopathologic Features in

Esophageal Carcinoma 11 Takuma Nomiya, Kenji Nemoto and Shogo Yamada

Chapter 3 Prognostic Indicators in Oral Squamous Cell Carcinoma 51

Márcio Diniz-Freitas, Eva Otero-Rey, Andrés Blanco-Carrión,Tomás García-Caballero, José Manuel-Gándara Rey and Abel García-

García

Chapter 4 Tumor-Targeting Non-Viral Gene Therapy for the Treatment of

Oral Cancer 95 Yoshiyuki Hattori and Yoshie Maitani

Chapter 5 New Diagnostic Imaging Modalities for Oral Cancers 125 Yasuhiro Morimoto, Tatsurou Tanaka, Izumi Yoshioka,

Yoshihiro Yamashita, Souichi Hirashima, Masaaki Kodama,Wataru Ariyoshi, Taiki Tomoyose, Norihiko Furuta, Manabu Habu,

Sachiko Okabe, Shinji Kito, Masafumi Oda, Hirohito Kuroiwa,

Nao Wakasugi, Tetsu Takahashi and Kazuhiro Tominaga

Chapter 6 The Role of the Percutaneous Endoscopic Gastrostomy in the

Management of Head and Neck Malignancy 155 CME Avery

Chapter 7 The Biomechanical Basis for Internal Fixation of the Radial

Osteocutaneous Donor Site 183 CME Avery


8/320

Contents vii

Chapter 8 The Current Role of Prophylactic Internal Fixation of the Radial

Osteocutaneous Donor Site

195

CME Avery

Chapter 9 Cytologic Diagnosis of Oral Malignancies: Scope and Limitations 211

Dilip K. Das Chapter 10 Benign and Malignant Tumors Occurring in the Pterygopalatine

Fossa and Adjacent Structures of the Pterygopalatine Fossa: Recent

Advances of Diagnosis and Surgical Management 229 Xin-Chun Jian

Chapter 11 Molecular Aspects of Oral Cancer: the Role of Phase I and II

Biotransformation Enzymes in Carcinogenesis 247 Karin Soares Gonçalves Cunha and Dennis de Carvalho Ferreira

Chapter 12 TP53 Mutation, c-myc Amplification and Squamous Cell CarcinomaRecurrence 263 J. Seoane, P. Varela-Centelles, M.A. Romero , A. De la Cruz, F. Barros,

L. Loidi and J.L. López Cedrún

Chapter 13 Recent Advances and Future Prospects Upon the Arterial Framework

of the Face and Related Applications for Facial Flaps 275 Egidio Riggio

Index 285


9/320


10/320

PREFACE

Oral cancer is any cancerous tissue growth located in the mouth. It may arise as a

primary lesion originating in any of the oral tissues, by metastasis from a distant site oforigin, or by extension from a neighboring anatomic structure, such as the nasal cavity or the

maxillary sinus. Oral cancers may originate in any of the tissues of the mouth, and may be of

varied histologic types: teratoma, adenocarcinoma derived from a major or minor salivary

gland, lymphoma from tonsillar or other lymphoid tissue, or melanoma from the pigment

producing cells of the oral mucosa. Far and away the most common oral cancer is squamous

cell carcinoma, originating in the tissues that line the mouth and lips. Oral or mouth cancer

most commonly involves the tissue of the lips or the tongue. It may also occur on the floor of

the mouth, cheek lining, gingiva (gums), or palate (roof of the mouth). Most oral cancers look

very similar under the microscope and are called squamous cell carcinoma. These aremalignant and tend to spread rapidly. This new book presents important research from around

the world.

Chapter 1 - Objective: Photodynamic therapy (PDT) is a promising cancer treatment in

which a photosensitizing drug accumulates in tumors and is subsequently activated by visible

light of an appropriate wavelength matched to the absorption. The advantages of this method,

as compared to other conventional cancer treatment modalities, are its low systemic toxicity

and its ability to destroy tumors selectively. 5-aminolevulinic acid (ALA)-induced

protoporphyrin-IX (PpIX) has been used as a photosensitizer in PDT for oral cancer, which

advantage is low side effect compared to other photosensitizer. This study investigates the

optimal method of administrating ALA by analyzing PpIX fluorescence in tongue tumor

tissue.

Methods: PpIX intensities in the mouse (C3H) transplanted tongue cancer (NR-S1) were

compared with those in normal tongue after intraperitoneal (i.p.), oral (p.o.), or topical

administration of ALA. Tongues were sampled at various times after ALA administration.

PpIX intensities were obtained from frozen sections of each sample by using a

spectrophotometer.

Results: PpIX intensity in the tumor group peaked at 3 h after the i.p. and 5 h after the

p.o. administration of ALA, and these levels were about twice as high as those in the normal

group. Maximum PpIX accumulation in the tongue tumor tissue was seen at 5 h after the oral


11/320

Alexios P. Nikolakakosx

administration of ALA. In contrast, the topical administration of 20% ALA cream was

associated with the lowest PpIX accumulation in the tumor throughout the experiments.

Conclusion: Based on these results, most effective administration route of ALA was oral

administration and 5 h after administration was regarded to be the optimal time for light

irradiation in ALA-PDT.

Chapter 2 - The clinical characteristics and radiosensitivity of esophageal cancer differ

individually, even in individuals with the same histopathological type. Several investigators

have reported that prognosis of patients with esophageal carcinoma differs according to its

macroscopic appearance, and it has been shown that macroscopically infiltrative type (like

scirrhous type in gastric cancer) is radioresistant and that its prognosis is extremely poor

compared to that of macroscopically localized type. The major factors that are thought to

have a potent impact on radiosensitivity of a tumor are cell proliferation activity, tumor

oxygenation, genetic repair, and intrinsic radiosensitivity.

In our study, Ki67, CD34, vascular endothelial growth factor (VEGF), thymidine

phosphorylase (TP) and metallothionein (MT) expressions and microvascular density wereevaluated using surgically resected esophageal squamous cell carcinomas without

preoperative treatment. Microvascular density (MVD) was evaluated in different ways:

average-MVD was estimated as an index of tumor oxygenation, and highest-MVD was

estimated as an index of the most active neovascularization in the tumor.

In the analysis of proliferation activity (Ki67 labeling index), proliferation activity of the

radiosensitive group of esophageal carcinomas was higher than that of the radioresistant

group of esophageal carcinomas. In the analysis of microvascular density, average-MVD of

macroscopically infiltrative type was significantly lower than that of localized type, whereas

highest-MVD of macroscopically infiltrative type was significantly higher than that oflocalized type. The VEGF expression level of infiltrative type was significantly higher than

that of localized type. A significant positive correlation was found between highest

microvascular density and VEGF expression, and a borderline significant negative correlation

was found between average microvascular density and expression of VEGF. TP expression

showed a positive correlation with highest-MVD, but the correlation was not as strong as that

of VEGF expression. In the analysis of MT, which is recognized as a protein that has a

radioprotective effect, expression of MT was not increased in esophageal carcinoma of the

radioresistant group. Metallothionein expression was increased in the radiosensitive group.

Furthermore, expression of MT was not increased in preoperatively treated esophageal

carcinomas. These results suggested that MT does not have a great impact on clinical

radiosensitivity in esophageal carcinoma and also suggested that MT expression is not

induced by therapeutic irradiation or anticancer agents.

The results suggest that radioresistant type is poorly oxygenated by low average-MVD,

includes a large amount of hypoxic fraction that is refractory to treatment, shows induction of

angiogenic factors and activated neovascularization, and has a high rate of hematogenous

metastasis. Tumor oxygenation and presence of a hypoxic fraction seem to have great

importance for curability of esophageal carcinoma compared to various other factors the

authors have investigated.

Chapter 3 - Every year, more than 300,000 new cases of oral cancer are diagnosed

worldwide. Oral squamous cell carcinomas (OSCCs) make up about 90 - 95% of these cases.


12/320

Preface xi

Despite intensive research into treatment modalities for oral cancer, the 5-year survival rate

has shown little improvement in recent decades. One of the reasons for this is that the TNM

classification system (the conventional basis for treatment decisions, in conjunction with

histological tumor grade) has proved not to be a consistently good predictor of prognosis.

There is thus a pressing need for research into new prognostic indicators, with the aim of

enabling the evaluation of the biological aggressiveness of each patient's particular tumor/s.

In recent decades, considerable research effort has been dedicated to the identification of

new markers of OSCC, with the aim of better predicting tumor behavior and clinical course.

Certainly, an improved knowledge of the different biological mechanisms participating in

carcinogenesis, as well as of cell proliferation, apoptosis, tumor growth and tumor invasive

capacity, may assist individual diagnosis, and help in the development of new treatment

strategies.

The aim of the present chapter is to briefly review the use of tumor markers for

prediction of the biological behavior of OSCCs. The review is divided into three parts,

considering first clinical markers, then histological markers, and finallyimmunohistochemical markers.

Chapter 4 - Despite advances in surgery, radiotherapy, and chemotherapy, the survival of

patients with oral squamous cell carcinoma has not significantly improved over the past

several decades. Gene therapy has the potential for the treatment of oral cancer. Cancer gene

therapy is currently being met with the development of non-viral vectors, because non-viral

vectors have a much lower potential for an adverse inflammatory or immune reaction,

compared with viral vectors. For gene delivery, oral cancer is a particular appropriate target

since it can be applied by direct injection. Also since folate and transferrin receptors are

frequently overexpressed on oral tumors such as nasopharyngeal tumor and head and neck ofsquamous cell carcinoma, folic acid and transferrin have been utilized as a ligand for tumor-

targeting gene delivery. Non-viral vectors conjugated to these ligands have been used as

carriers of therapeutic DNA to targeted oral tumor. The strategies are used for inactivation of

oncogene expression, introduction of tumor suppressor genes, and introduction of a gene that

enable to a prodrug to be activated into an active cytotoxic drug. In this review, the authors

outline tumor-targeting liposome and lipid-based nanoparticle vectors, and discuss the

effectiveness as these non-viral vectors for DNA transfection and for gene therapy to treat

human oral tumors.

Chapter 5 - This article reviews the use of imaging modalities; both commonly used and

recently introduced, to evaluate oral cancers and their lymph node metastases. Magnetic

resonance images (MRI) and X-ray computed tomography (CT) images are used to determine

the size, invasive area, and possible pathology of primary cancers. In addition, the two

modalities are useful for staging and detecting clinically occult lymph node metastases at

different levels of the neck. In particular, a follow-up MR examination method, dynamic MR

sialography, for patients with xerostomia after radiation therapy is introduced, and the use of

fusion images of the tumors and vessels using three-dimensional fast asymmetric spin-echo

(3D-FASE) and MR angiography is discussed. Furthermore, ultrasound imaging (US), in

addition to its use for staging and detecting clinically occult lymph node metastases, plays an

important role in confirming intra-operative surgical clearance of tongue carcinomas. In

addition, the role of US-guided, fine-needle aspiration biology is also reviewed. Finally, the


13/320

Alexios P. Nikolakakosxii

role and limitations of fusion images obtained from positron emission tomography (PET) and

CT (PET-CT), which are currently used worldwide, are discussed.

Chapter 6 - This chapter reviews the role of the percutaneous endoscopic gastrostomy

(PEG) for providing nutritional support in the management of oral cancer. An assessment of

the current use of the PEG technique is based on an analysis of the prospective operating

series of the author.

Insertion of a PEG was attempted on 200 occasions, mainly for malignancy of the oral

cavity but also the oropharynx, and some benign pathology and trauma. Seventy-six percent

(152/200) of gastrostomies were inserted at the time of definitive surgical treatment and

19.5% (39/200) were inserted at an examination under anaesthesia, often prior to

radiotherapy.

Five percent (10/200) of procedures had significant endoscopic findings including one

synchronous malignancy. The rate of successful insertion was 97% (194/200). The incidence

of minor and major complications was 12.5% (25/200) and 3% (6/200) respectively. There

was no procedure related mortality. The overall 30-day mortality rate was 7% (10/200)including deaths from terminal disease. Those at increased risk of death were 65 years and

over (P=0.005). The median PEG duration was 287 (SE 37) days. Duration was significantly

longer for stage T3-4 tumours (P=0.01), N1 or greater neck disease (P=0.02), following

surgery with radiotherapy when compared to surgery alone (P


14/320

Preface xiii

reinforcement with different types of 3.5mm plate. The plate was placed in either the anterior

(over the defect) or posterior (on the intact cortex) position.

An osteotomised bone was significantly weaker than an intact bone. A plate in either the

anterior or posterior position significantly strengthened an osteotomised bone. The dynamic

compression plate was the strongest reinforcement in both torsion and bending. In torsion the

mean strength of the intact bone was 45% greater than after osteotomy (P=0.02). The

reinforced bone was on average 61% stronger than the unreinforced bone (P


15/320

Alexios P. Nikolakakosxiv

is significant peripheral vascular disease or poor general health that will be exacerbated by

use of an alternative donor site.

Chapter 9 - Cancer of mouth and pharynx is one of the ten most common cancers in the

world. Detection of a precancerous or cancerous lesion at an early stage is an important factor

to improve 5-year survival rate of oral cancer. A comprehensive physical examination aided

by imaging techniques like computed tomography (CT), and magnetic resonance imaging

(MRI) are the standard evaluation tools in patients with oral, and pharyngeal neoplasms.

Although surgical biopsy and histopathology is considered gold standard for diagnosing the

oral lesions, it is impractical to routinely subject large number of patients to biopsy. Whereas

oral exfoliative cytology is a useful, economical and practical tool in the diagnosis of oral

dysplasia and carcinoma involving cheek, lip and tongue, similar role is played by fine needle

aspiration (FNA) cytology for minor salivary gland tumors and other solid neoplasms of the

palate, cheek and pharyngeal areas. By brush cytology a spectrum of oral lesions including

dysplasia, carcinoma in situ, occult and clinically evident squamous cell carcinoma can be

diagnosed. FNA cytology, which collects samples from areas difficult to reach by surgical biopsy, can differentiate benign from malignant tumors and classify them into subtypes.

Whereas pleomorphic adenoma is a common benign tumor, adenoid cystic carcinoma,

mucous cell carcinoma, acinic cell carcinoma, malignancy in pleomorphic adenoma, and

polymorphous low-grade carcinoma are the malignant neoplasms detected in the minor

salivary glands. The other oral neoplasms detected by FNA are non-Hodgkin lymphomas,

and some rare primary malignancies like sarcomas and chordoma. Metastatic lesions in oral

cavity too have been diagnosed by FNA cytology. The efficacy of brush cytology in detection

of oral squamous cell carcinoma is very high in majority of reports, which is as follows:

sensitivity (84.4 ± 9.97%), specificity (78.6 ± 29.36%), positive predictive value (71.4 ±31.39%), and negative predictive value (83.0± 16.40%). The sensitivity, specificity, and

diagnostic accuracy of FNA cytology for oral malignancies are also high. However, false

negative reports are possible with the oral brush cytology technique and some palatal salivary

gland tumors are difficult to diagnose by FNA cytology. In difficult situations, ancillary

techniques such as cytomorphometry, DNA-cytometry, immunocytochemistry, and molecular

tools act as valuable adjunct to cytodiagnostic techniques.

Chapter 10 - Surgery in the pterygopalatine fossa region presents anatomic and surgical

problems related to the difficulty of access. When a tumor in the pterygopalatine fossa

involves the maxilla and extends into the maxillary sinus and a tumor of the deep lobe of the

parotid gland extends into the pterygopalatine foss, extensive resection is often necessary.

Because of this, there has been a tendency either not to operate on these cases at all or else to

carry out simply a partial or piecemeal removal. The current underlying principle of skull

base approaches is to minimize brain retraction while maximizing skull base visualization.

This concept facilitates three-dimensional tumor resection, tumor margin verification, and

functional reconstruction with appropriate esthetic concerns. Current many approaches have

been used for the tumor of the middle skull base or the pterygopalatine fossa.

With advancements in imaging, diagnostic technology, diagnostic pathology, surgical

technology and instrumentation, reconstructive techniques, the surgery of the lateral cranial

base or the middle cranial base is now receiving significant attention and interest. It is

purpose of this paper to provide readers with an overall review of benign and malignant


16/320

Preface xv

tumors occurring in the pterygopalatine fossa and adjacent structures of the pterygopalatine

foss: Recent advances of diagnosis and surgical management.

Chapter 11 - Oral cancer is the most common malignant neoplasm of the head and neck

and over half of the people who develop this cancer die within five years after the diagnosis.

Carcinogenesis is a highly complex process involving both environmental, mainly tobacco

and alcohol use, and inherited risk factors. In recent years, inter-individual genetic

differences and individual susceptibility to human cancer triggered by environmental

exposures has been studied. This environment-gene interaction in carcinogenesis is well

reflected by phase I and II enzymes that are involved in the metabolism of carcinogens.

Cytochrome P450 family of enzymes (CYP), involved in phase I, converts many carcinogens

into DNA-binding metabolites in target cells and can modulate intermediate effect markers

such as DNA-adducts. Phase II enzymes, including glutathione S-transferase (GST), N-

acetyltransferase (NAT) and others, play important roles in protecting cells from DNA

damage by carcinogens and reactive oxygen species. Genetic alterations of these two classes

of enzymes have been considered as risk modifiers of some major tobacco-related cancers,including oral cancer. The aim of this chapter is to review the molecular aspects of oral

cancer, emphasizing the role of phase I and II enzymes in oral carcinogenesis. Propositions

for further researches are highlighted.

Chapter 12 - Purpose: to investigate TP53 mutation and c-myc amplification as markers

for tumour aggressiveness in terms of tumour recurrence in OSCCs.

Methods and materials: Thirty one incident cases of oral squamous cell carcinomas were

studied for tumour relapse. The variables considered were demographic, clinical, pathological

and genetic.

Results: the mean age of 62.09 years (range 36 to 88). Seventeen patients (54.8%) weresmokers. The tongue was the main affected area (54.8%). No distant metastases could be

identified. Most patients were at early stages of the disease with moderately differentiated

tumours and of grade I in Anneroth’s malignancy scale. The oncogene study showed

abnormalities in both TP53 (6/31; 19.2%) and c-myc (4/31; 12.9%), that distributed as

follows: TP53+/c-myc+ (n=1; 3.2 %); TP53+/c-myc- (n=5; 16.1%); TP53-/c-myc+ (n=3; 9.7

%); TP53-/c-myc- (n=21; 67.7%). TP53 mutations were significantly more frequent in

advanced stages. Statistically significant differences in node status were identified in terms of

oncogene alterations. Multivariate Cox regression analysis recognized prognostic value for

recurrence for alterations of TP53 and c-myc (p


17/320

Alexios P. Nikolakakosxvi

temporal artery) or the terminal branches of the frontal terminal branch, from the variants of

the terminal facial artery and a definite collateral named cutaneous zygomatic branch, or from

the submental artery. The up-to-date research embraces the study of the cutaneous perforators

of the face. Relevant anterograde or reverse flaps, axial or perforator flaps, and monolayered

or multilayered composite flaps are discussed as current, original or still imaginative chances.

Moreover, for the realization of totally new flaps in the field of compound facial

reconstruction, clinical research efforts should tend to merge with the future perspective of

bone and soft-tissue engineering research.


18/320

In: Oral Cancer Research Advances ISBN 978-1-60021-864-4

Editor: Alexios P. Nikolakakos, pp. 1-10 © 2007 Nova Science Publishers, Inc.

Chapter 1

EFFECTIVE ADMINISTRATION METHODS

OF 5-AMINOLEVULINIC ACID AS A

PHOTOSENSITIZER IN PHOTODYNAMICTHERAPY FOR TONGUE TUMOR

Toshiyuki Ogasawara1,

, Norio Miyoshi 2 , Kazuo Sano

1 ,

Hidetaka Kinoshita1 , Tetsushi Yamada

1 , Toru Ogawa

1 ,

Kazuki Miyauchi1 and Yoshimasa Kitagawa

3

1

Division of Dentistry and Oral Surgery, Department of Sensory and LocomotorMedicine,

2Division of Tumor Pathology, Department of Pathological Sciences, School

of Medicine, Faculty of Medical Sciences, University of Fukui, Fukui 910-1193, Japan;3Oral Diagnosis and Oral Medicine, Department of Oral Pathobiological Science,

Graduate School of Dental Medicine, Hokkaido University, Sapporo 060-8586, Japan.

ABSTRACT

Objective: Photodynamic therapy (PDT) is a promising cancer treatment in which a photosensitizing drug accumulates in tumors and is subsequently activated by visible

light of an appropriate wavelength matched to the absorption. The advantages of this

method, as compared to other conventional cancer treatment modalities, are its low

systemic toxicity and its ability to destroy tumors selectively. 5-aminolevulinic acid

(ALA)-induced protoporphyrin-IX (PpIX) has been used as a photosensitizer in PDT for

oral cancer, which advantage is low side effect compared to other photosensitizer. This

study investigates the optimal method of administrating ALA by analyzing PpIX

fluorescence in tongue tumor tissue.

∗

Correspondence concerning this article should be addressed to: Toshiyuki Ogasawara, Division of Dentistry and

Oral Surgery, Department of Sensory and Locomotor Medicine, School of Medicine, Faculty of Medical

Sciences, University of Fukui, 23-3 Matsuokahimoaizuki, Eiheiji, Fukui 910-1193, Japan. Tel: 81-776-61-

3111(ex2409); Fax: 81-776-61-8128; E-mail: [email protected].


19/320

Toshiyuki Ogasawara, Norio Miyoshi, Kazuo Sano et al.2

Methods: PpIX intensities in the mouse (C3H) transplanted tongue cancer (NR-S1)

were compared with those in normal tongue after intraperitoneal (i.p.), oral (p.o.), or

topical administration of ALA. Tongues were sampled at various times after ALA

administration. PpIX intensities were obtained from frozen sections of each sample by

using a spectrophotometer.

Results: PpIX intensity in the tumor group peaked at 3 h after the i.p. and 5 h after

the p.o. administration of ALA, and these levels were about twice as high as those in the

normal group. Maximum PpIX accumulation in the tongue tumor tissue was seen at 5 h

after the oral administration of ALA. In contrast, the topical administration of 20% ALA

cream was associated with the lowest PpIX accumulation in the tumor throughout the

experiments.

Conclusion: Based on these results, most effective administration route of ALA was

oral administration and 5 h after administration was regarded to be the optimal time for

light irradiation in ALA-PDT.

Keywords: 5-aminolevulinic acid, protoporphyrin-IX, photodynamic therapy, tongue cancer,

spectroscopy, pharmacokinetic

INTRODUCTION

Surgery with radiotherapy and / or chemotherapy has been used as the conventional

treatment for tongue cancer. However, this treatment causes cosmetic and functional

disturbances, especially in the head and neck region.

Photodynamic therapy (PDT) is a promising cancer treatment in which a photosensitizing

drug accumulates in tumors and is subsequently activated by visible light of an appropriate

wavelength matched to the absorption [1]. The advantages of this method, as compared to

other conventional cancer treatment modalities, are its low systemic toxicity and its ability to

destroy tumors selectively [2]. Photofrin is the most widely used photosensitizer in clinical

PDT trials and is the only agent that has been approved for cancer treatment in many

countries. However, photofrin remains in the skin and causes photosensitivity lasting several

weeks, and the tumor selectivity of this agent is poor [3]. 5-aminolevulinic acid (ALA) is a

precursor of protoporphyrin IX (PpIX) in the biosynthetic pathway for heme, and PpIX is an

efficient photosensitizer. Today ALA–PDT is successfully used for the treatment of a variety

of neoplastic and nonneoplastic diseases [4]. ALA- derived PpIX can be cleared from the body within 24-48 h after systemic ALA administration [5], and because of this rapid

clearance, ALA-based PDT would reduce the risk of prolonged skin phototoxicity [6].

The kinetics of ALA-induced PpIX production in different tissues has been studied,

typically by means of fluorescence spectroscopic techniques [7]. However, the relationship

between the PpIX fluorescent accumulation in oral tumor tissue and the ALA administration

methods has not been elucidated. This study investigated the optimal method for

administrating ALA in PDT by analyzing PpIX fluorescence in tongue tumor tissue.


20/320

Administration Methods of ALA in Tongue Tumors 3

MATERIALS AND METHODS

Animals and Tumors

Male C3H/HeNCrj mice, 6-8 weeks old, 22-26 g (Charles River, Osaka, Japan) were

used in all experiments.

An NR-S1 mouse squamous cell carcinoma [8] (National Institute of Radiological

Sciences, Chiba, Japan) was transplanted to the mouse tongue, and when the tumor reached a

size of at least 3mm x 3mm, the photosensitizer was administered. The photosensitizer was

also administered to normal mice as a control.

Chemicals and ALA Administration Route

ALA was obtained as a hydrochloride in 98.0% pure powder from Cosmo Oil (Tokyo,

Japan).

In the intraperitoneal ALA administration group, ALA was freshly dissolved in 0.2 ml of

saline and injected at a dose of 250 mg/kg or 500 mg/kg.

In the oral ALA administration group, animals were given 250 mg/kg or 500 mg/kg of

ALA freshly dissolved in 0.5 ml of saline by means of a gastric tube.

In the case of topical ALA administration, an oil-in-water emulsion containing 20% ALA

was freshly prepared prior to use. After topical administration of ALA cream to the tongue,

animals were maintained under deep anesthesia by pentobarbital sodium to in order to

prevent the ALA cream from being washed out or swallowed. Furthermore, two kinds ofALA ester derivative (ALA methyl ester and ALA pentyl ester; Cosmo Oil) were also

administrated and compared with topical application. These ALA ester derivatives, are more

lipophilic than ALA, and thus may penetrate more easily through the keratinized layer and

deeper into tumors than ALA itself [9].

Mice were killed at 1, 3, 4, 5, 6 and 8 h after ALA administration (n=4 animals/time

point), and mouse tongue samples were excised. Serial frozen sections (10μm-thick) of each

sample were prepared for exact histological localization and quantitative measurement of the

concentration of PpIX. PpIX localization was confirmed by fluorescence microscopy

(PROVIS-AV80type; Olympus, Tokyo, Japan) by comparing with a hematoxylin-eosin (H-E)-stained section. The wavelength width of the excitation filter was in the blue violet region

(400-440 nm), and the observation wavelength was more than 475 nm.

Quantitative Measurement of PpIX Fluorescence

ProtoporphyrinIX intensities in the mouse transplanted tongue cancer were compared

with those in the normal tongue after intraperitoneal, oral, or topical administration of ALA.

Levels of PpIX fluorescence were measured with a spectrophotometer [10].


21/320


Fluorescence emission spectra excited by 410 nm light were obtained from a total of 5

serial frozen sections (10μm-thick) from each sample by using a spectrophotometer (850

type; Hitachi, Tokyo, Japan) equipped with a holder for the particle sample.

The subtracted spectrum was obtained by subtracting the background spectrum from the

sample raw spectrum.

A typical fluorescence spectrum from a tumor showed prominent emission bands at

λ =635 nm and λ =705 nm, which corresponded to the standard PpIX spectrum (Figure 1).

The PpIX concentration (μM) was calculated from the fluorescence intensity at the 635

nm peak of the sample emission spectrum and a calibration curve of the known

concentrations of standard PpIX solution. Standard PpIX aqueous solution was prepared with

phosphate-buffered saline solution, cationic surfactant, acetyl-trimethyl-ammonium-bromide

and PpIX.

Figure 1. Fluorescence emission spectra excited by 410 nm light are obtained from a total of 5 serialfrozen sections (10μm-thick) from each sample by using a spectrophotometer. The no.3 subtracted

spectrum was obtained by subtracting the no.2 background spectrum from the no.1 sample raw

spectrum. In addition, we confirmed that the no.3 subtracted spectrum pattern corresponded to the no.4

standard PpIX spectrum.

Statistics

Groups of normally distributed data were compared using Student`s t-test, while the non-

parametric Mann-Whitney test was otherwise employed. Values of


22/320


R ESULTS

The fluorescence microscopic image showed that the red fluorescence emission of PpIX

was distributed strongly and homogeneously in the tongue tumor tissue at 5 h after oral

administration of ALA. However, PpIX accumulation was not seen in the necrotic area of the

tumor tissue. In addition, there was very weak PpIX accumulation in the normal lingual

muscle after administration of ALA.

Figure 2. The PpIX concentration (μM) was calculated from the fluorescence intensity at the 635 nm

peak of the subtracted spectrum and a calibration curve of the known concentrations of standard PpIX.

Figure 3. Fluorescence image and corresponding HE-stained image of tongue tumor tissue at 5 h after

oral administration of ALA.

The tumor group showed constantly higher PpIX intensities than the normal group

throughout the experiments following the i.p. and p.o. administration of ALA. PpIX intensity

in the tumor group peaked at 3 h after the i.p. and 5 h after the p.o. administration of ALA,

and these peak values were about twice as high as those in the normal group. However, the


23/320


PpIX intensitiy in the tumor group was not enhanced by an increase in the administrated dose

of ALA from 250 mg/kg to 500 mg/kg (Figures 4 and 5). Maximum PpIX accumulation in

the tongue tumor tissue was seen at 5 h after the oral administration of ALA (Figure 6). In

contrast, the topical administration of 20% ALA cream was associated with the lowest PpIX

accumulation in the tumor throughout the experiments (Figure 7). Furthermore, the topical

administration of 20% ALA ester derivatives cream (ALA methyl ester and ALA pentyl

ester) also resulted in low PpIX accumulation in the tumor, which was not different from the

case of topical administration of 20% ALA cream.

Figure 4. PpIX intensity in tongue tissue after intraperitoneal administration of ALA.

Figure 5. PpIX intensity in tongue tissue after oral administration of ALA.


24/320


Figure 6. PpIX intensity in tongue tumor tissue after various types of ALA administration.

Figure 7. PpIX intensity in tongue tissue after topical administration of ALA.

DISCUSSION

If the photosensitizer that is administered before light illumination accumulates more

highly in tumor tissue, the efficacy of PDT for cancer might be improved. Although photofrinshould be used only via intravenous administration, ALA can be used via various


25/320


administration routes. There have been numerous studies on ALA-induced PpIX fluorescence

after a variety of administration routes in various organs. Systemic (intravenous or oral)

ALA-based PDT has also been reported for treatment of oral neoplastic lesions [5,11].

However, the optimal administration method of ALA in PDT for oral cancer is still not

established. The present study was carried out to determine the most effective route and

optimal timing of ALA administration with respect to subsequent therapeutic illumination.

In most studies, the techniques used were based on a noninvasive method using a

spectrofluorometer or the chemical technique of high pressure liquid chromatography

(HPLC) to measure in vivo PpIX fluorescence after administration of ALA.

Spectrofluorometry is simple and noninvasive, but can detect only surface emission of the

tissue. HPLC can measure the whole tissue, but the resulting values are an average for the

whole tissue (more than 1 g) and have no relation to the histopathological findings.

Furthermore, the techniques required for this method are complicated [10]. We confirmed the

direct detection the PpIX concentrations from the frozen section always in combination with

histological staining using cryosamples.Previous studies have shown that the peak of PpIX fluorescence intensity varied between

1 and 6 h after ALA administration in different tissues [5,12,13]. Our present results showed

that PpIX intensity in the tongue tumor tissue peaked at 3 h after the intraperitoneal (i.p.)

administration of ALA. Another study has also shown that PpIX fluorescence in rat tongue

cancer reached a maximum intensity at 3 h after ALA i.p. administration [13]. However, in

this study, the maximum PpIX accumulation in the tongue tumor tissue was confirmed at 5 h

after oral administration of ALA. Although the reason for this finding is unclear, Mustajoki et

al. [14] have shown that a high serum ALA level can be achieved in a human volunteer by

continuous enteral infusion of ALA solution. Loh et al. [15] reported that the temporalfluorescence kinetics after oral administration were comparable with that after intravenous

injection in the stomach, colon and bladder mucosa of normal rats. Oral administration is

considered to be simpler, and it does not require full buffering. ALA can be undertaken by

patients themselves, prior to therapy and without supervision [15]. The results of the present

study suggested that oral administration was the most effective administration method in

ALA-PDT for oral cancer.

Furthermore, there was no obvious difference of PpIX intensity between 250 mg/kg and

500 mg/kg after both i.p. or p.o. administration of ALA. Accumulation of PpIX in tongue

tumor tissues reaches a plateau after administering at least 250 mg/Kg doses of ALA. Ma et

al. [13] reported that early malignant lesions in rat tongue showed complete response to the

i.p. administration of ALA-based PDT at both 250 mg/Kg and 1000 mg/Kg. These results

suggested that it is not necessary to administer a greater amount of ALA in order to achieve

sufficiently high PpIX levels suitable for PDT in oral cancer.

Topical ALA-based PDT has been widely used in treating neoplastic lesions of the skin

and bladder [4], because local administration of ALA might increase the PpIX concentration

in the tumor without unwanted general side effects. It is known that topical application of an

oil-in-water emulsion of ALA on the skin lesion can permit penetration of ALA into the

lesion and allow synthesis of PpIX [16,17]. Recently, in cases of oral cancer, topical

administration of ALA as a rinsing solution has also been tried for ALA-photodynamic

diagnosis. However, because there has been no report on the use of topical administration


26/320


ALA-PDT for oral cancer, we here tried topical administration of 20% ALA cream for tongue

tumor. Our results showed that PpIX intensity in the tongue tumor after topical administration

of ALA cream was not enhanced compared with that in the normal tongue. Several ALA

esters have been synthesized, and are more lipophilic than ALA. This higher lipophilicity

might result in better penetration into the skin, higher PpIX levels and a more uniform and

deeper PpIX distribution [18]. Therefore, we attempted to apply the two kinds of ALA esters

for topical administration. However, ALA esters also did not enhance the PpIX intensity in

the tongue tumor tissue. Although the reason for this result is unclear, the neutral pH of saliva

might cause the immediate degeneration of ALA in the oral mucosa [15,18]. Based on these

results, 5 h after oral administration of ALA was regarded to be the optimal time for light

irradiation in ALA-PDT.

ACKNOWLEDGEMENT

This work was supported by a Grant-in-Aid for Scientific Research (C) (15592104) from

Japan Society for the Promotion of Science.

R EFERENCES

[1] Date M, Sakata I, Fukuchi K et al. (2003). Photodynamic therapy for human oral

squamous cell carcinoma and xenografts using a new photosensitizer, PAD-S31. Lasers

Surg Med 33: 57-63.

[2] Gaullier JM, Berg K, Peng Q et al. (1997). Use of 5-aminolevulinic acid esters to

improve photodynamic therapy on cells in culture. Cancer Res 57 : 1481-1486.

[3]

Peng Q, Berg K, Moan J et al. (1997). 5-aminolevulinic acid-based photodynamic

therapy: principles and experimental research. Photochem Photobiol 65: 235-251.

[4]

Peng Q, Warloe T, Berg K et al. (1997). 5-Aminolevulinic acid –based photodynamic

therapy. Cancer 79: 2282-2308.

[5] Grant WE, Hopper C, MacRobert AJ et al. (1993). Photodynamic therapy of oral

cancer:photosensitization with systemic aminolevulinic acid. Lancet 342: 147-148.

[6]

Webber J, Kessel D, Fromm D (1997). Side effects and photosensitization of humantissue after aminolevulinic acid. J Surg Res 68 : 31-37.

[7] Stolic S, Tomas SA, Roman-Gallegos E et al. (2002). Kinetic study of δ-Ala induced

porphyrins in mice using photoacoustic and fluorescence spectroscopies. J Photochem

Photobiol B: Biol 68 : 117-122.

[8]

Usui S, Urano M, Koike S et al (1976). Effect of PSK, a protein polysaccharide, on

pulmonary metastasis of C3H mouse squamous cell carcinoma. J Natl Cancer Inst 56 :

185-187.

[9] Juzenas P, Shafaei S, Moan J et al. (2002). Proitoporphyrin IX fluorescence kinetics in

UV-induced tumours and normal skin of hairless mice after topical application of 5-aminolevulinic acid methyl ester. J Photochem Photobiol B: Biol 67 : 11-17.


27/320


[10]

Miyoshi N, Ogasawara T, Nakano K et al. (2004). In light of recent developments,

application of fluorescence spectral analysis in tumor diagnosis. Appl Spectrosc Rev 39:

437-455.

[11]

Fan KN, Hopper C, Speight PM et al. (1996). Photodynamic therapy using 5-

aminolevulinic acid for premalignant and malignant lesions of oral cavity. Cancer 78 :

1374-1383.

[12] Henderson BW, Vaughan L, Bellnier DA et al. (1995). Photosensitization of murine

tumor, vasculature and skin by 5-aminolevulinic acid-induced porphyrin. Photochem

Photobiol 62: 780-789.

[13]

Ma G, Ikeda H, Inokuchi T et al. (1999). Effect of photodynamic therapy using 5-

aminolevulinic acid on 4-nitroquinoline-1-oxide-induced premalignant and malignant

lesions of mouse tongue. Oral Oncol 35: 120-124.

[14]

Mustajoki P, Timonen K, Gorchein A et al. (1992). Sustained high plasma 5-

aminolevulinic acid concentration in a volunteer: no porphyric symptoms. Euro J Clin

Invest 22: 407-411.[15] Loh CS, MacRobert AJ, Bedwell J et al. (1993). Oral versus intravenous administration

of 5-aminolaevulinic acid for photodynamic therapy. Br J Cancer 68 : 41-51.

[16] Kennedy JC, Pottier RH (1992). Endogeneous protoporphyrin IX, a clininically useful

photosensitizer for photodynamic therapy. J Photochem Photobiol B 14: 275-292.

[17] Szeimies RM, Sassy T, Landthaler M (1994). Penetration potency of topical applied 5-

aminolevulinic acid for photodynamic therapy of basal cell carcinoma. Photochem

Photobiol 59: 73-76.

[18]

van den Akker JTHM, Lani V, Star WM et al (2000). Topical application of 5-

aminolevulinic acid hexyl ester and 5-aminolevulinic acid to normal nude mouse skin:differences in protoporphyrin IX fluorescence kinetics and the role of the stratum

corneum. Photochem Photobiol 72: 681-689.


28/320

In: Oral Cancer Research Advances ISBN 978-1-60021-864-4

Editor: Alexios P. Nikolakakos, pp. 11-50 © 2007 Nova Science Publishers, Inc.

Chapter 2

R ELATIONSHIPS BETWEEN BIOLOGICAL AND

CLINICOPATHOLOGIC FEATURES IN

ESOPHAGEAL CARCINOMA

Takuma Nomiya, Kenji Nemoto and Shogo YamadaDepartment of Radiation Oncology, Yamagata University School of Medicine, Japan.

ABSTRACT

The clinical characteristics and radiosensitivity of esophageal cancer differ

individually, even in individuals with the same histopathological type. Several

investigators have reported that prognosis of patients with esophageal carcinoma differs

according to its macroscopic appearance, and it has been shown that macroscopically

infiltrative type (like scirrhous type in gastric cancer) is radioresistant and that its

prognosis is extremely poor compared to that of macroscopically localized type. The

major factors that are thought to have a potent impact on radiosensitivity of a tumor are

cell proliferation activity, tumor oxygenation, genetic repair, and intrinsic

radiosensitivity.

In our study, Ki67, CD34, vascular endothelial growth factor (VEGF), thymidine

phosphorylase (TP) and metallothionein (MT) expressions and microvascular density

were evaluated using surgically resected esophageal squamous cell carcinomas without

preoperative treatment. Microvascular density (MVD) was evaluated in different ways:

average-MVD was estimated as an index of tumor oxygenation, and highest-MVD was

estimated as an index of the most active neovascularization in the tumor.

In the analysis of proliferation activity (Ki67 labeling index), proliferation activity of

the radiosensitive group of esophageal carcinomas was higher than that of the

radioresistant group of esophageal carcinomas. In the analysis of microvascular density,

average-MVD of macroscopically infiltrative type was significantly lower than that of

localized type, whereas highest-MVD of macroscopically infiltrative type was

significantly higher than that of localized type. The VEGF expression level of infiltrativetype was significantly higher than that of localized type. A significant positive

correlation was found between highest microvascular density and VEGF expression, and


29/320

Takuma Nomiya, Kenji Nemoto and Shogo Yamada12

a borderline significant negative correlation was found between average microvascular

density and expression of VEGF. TP expression showed a positive correlation with

highest-MVD, but the correlation was not as strong as that of VEGF expression. In the

analysis of MT, which is recognized as a protein that has a radioprotective effect,

expression of MT was not increased in esophageal carcinoma of the radioresistant group.

Metallothionein expression was increased in the radiosensitive group. Furthermore,

expression of MT was not increased in preoperatively treated esophageal carcinomas.

These results suggested that MT does not have a great impact on clinical radiosensitivity

in esophageal carcinoma and also suggested that MT expression is not induced by

therapeutic irradiation or anticancer agents.

The results suggest that radioresistant type is poorly oxygenated by low average-

MVD, includes a large amount of hypoxic fraction that is refractory to treatment, shows

induction of angiogenic factors and activated neovascularization, and has a high rate of

hematogenous metastasis. Tumor oxygenation and presence of a hypoxic fraction seem

to have great importance for curability of esophageal carcinoma compared to various

other factors we have investigated.

1. BACKGROUND AND EPIDEMIOLOGY

Esophageal cancer is a malignancy that has an extremely poor prognosis. Esophageal

carcinoma arises from squamous cells of the esophagus. Squamous cell carcinoma accounts

for most esophageal malignancies, and adenocarcinoma is the second-most frequently

occurring esophageal malignancy. The proportions of histological type differ according to

country and race. A recent survey has shown that the ratios of squamous cell carcinoma and

adenocarcinoma in esophageal malignancies in North America are 50-60% and 40-50%,respectively, whereas the ratios of squamous cell carcinoma and adenocarcinoma in Japan are

more than 90% and less than 5%, respectively [1,2].

Smoking, alcohol, hot meals, having Barrett esophagus, and genetic inheritance are

thought to be the causes of esophageal carcinoma. The difference in the proportions of

squamous cell carcinoma and adenocarcinoma of the esophagus might be due to genetic

differences between races and to environmental factors, though the reasons have not been

clarified.

Recent clinical studies on esophageal squamous cell carcinoma have been shown that the

5-year survival rate of patients with esophageal squamous cell carcinoma regardless of stageis about 20-30%. Esophageal carcinoma has thus been a refractory disease despite recent

advances in multimodal treatments. The reasons for the extremely poor prognosis are that

esophageal carcinoma easily extends to the submucosal layer, results in wide lymphogenous

metastases from the neck to abdomen, easily invades adjacent critical organs, easily

debilitates the host due to esophageal stenosis and eating disorder, and is difficult to resect

completely. It has been shown that prognostic factors of gastro-intestinal malignancies

include depth of invasion, extent of lymphogenous metastases, presence of distant metastasis,

tumor length, site, age, and extent of lymphatic/ blood vessel invasion [3-5].

In case of gastro-intestinal malignancies, it is known that the clinical characteristics and

malignant potential of the tumor differ according to its macroscopic appearance. For

example, gastric cancer is roughly classified into localized type and invasive type and is


30/320

Relationships Between Biological and Clinicopathologic Features… 13

classified according to presence of ulceration [6]. In gastric cancer, it has been reported that

not only the abovementioned depth of invasion, extent of lymphogenous metastases, and

lymphatic/ blood vessel invasion but also macroscopic appearance greatly affect patients'

prognosis [3,4,7-11]. The prognosis of patients with macroscopically infiltrative type of

gastric cancer is generally regarded as poor, and the prognosis of patients with Borrmann's

type IV (scirrhous type, diffusely infiltrating type) is extremely poor [7,12,13]. These

findings suggest that the morphologic difference shows difference in biological features such

as frequency of metastasis, invasiveness, and refractoriness to treatment.

There are guidelines for treatment of esophageal carcinoma in Japan in which

macroscopic appearance is defined according to Borrmann's classification for gastric cancer

[14-16]. Many studies have shown that there is a difference in prognoses of patients with

esophageal carcinoma according to macroscopic type in Japan [5,13,17-21]. Similar to gastric

cancer, the prognosis of patients with macroscopically infiltrative type of esophageal

carcinoma is unfavorable, and the prognosis of patients with diffusely infiltrative type

(similar to Borrmann's type IV) is extremely poor [13].

Figure 1. A) Survival curves of patients with esophageal carcinoma treated by radiotherapy alone

according to macroscopic types (Stage II-III). The prognosis of patients with macroscopically

infiltrative type of esophageal carcinoma is significantly poorer than that of patients with localized type.

B) Response to radiotherapy alone. Response rate (CR+PR) of infiltrative type is also significantly

worse than that of localized type.

2. TREATMENT OUTCOME AND CLINICAL FEATURES

Figure 1A shows survival curves of patients with stage II-III esophageal squamous cell

carcinoma treated with radiotherapy alone in our institution from 1981 to 1991 (n=156; 144

males, 12 females; median age, 68.5 years; age range, 46-91 years) [22]. The prognosis of

patients with macroscopically infiltrative type of esophageal carcinoma was significantly

poorer than that of patients with macroscopically localized type (p


31/320


patients differ according to macroscopic appearance in esophageal carcinoma. It can be said

that macroscopically infiltrative type of esophageal carcinoma not only has an extremely poor

prognosis but is also more radioresistant than localized type. However, it is not clear whether

the unfavorable prognosis of infiltrative type is due to metastasis tendency, poor local

controllability, or both of these. Based on these differences in clinical features of esophageal

carcinoma, we investigated tumor biological differences from the viewpoint of biological

characteristics such as cell proliferation activity, microvascular density, activity of

angiogenesis, and intrinsic radiosensitivity.

3. SIGNIFICANCE OF PROLIFERATION ACTIVITY IN

MALIGNANCIES

In our studies, cell proliferation activity, microvascular density, expressions of

angiogenesis factors, and expressions of factors that affect intrinsic radiosensitivity were

evaluated using surgically resected esophageal carcinoma specimens, not biopsy specimens.

Ki67 labeling index was used for evaluation of cell proliferation activity of the tumor by

immunohistochemistry. Ki67 antigen is one of the proteins that are expressed in the nuclei of

proliferating cells [23]. The Ki67 antibody combines with a nuclear antigen that is present in

proliferating cells but absent in resting cells. According to results of past experiments, Ki67

nuclear antigen is present in S, G2 and M phases of the cell cycle but is absent in G0 phase

[24,25]. Ki67 positivity in G1 phase differs depending on the cell line. Ki67 antibody can

sensitively detect cells in the proliferating phase. The ratio of Ki67-positive cells in

malignant tissue indicates growth rate of the tumor, and Ki67 labeling index is widely used as

an index of proliferation activity of the tumor in clinical pathology.

Expression of Ki67 antigen in various tissue has been reported, and it has been shown

that Ki67 positivity of a malignant tumor is generally higher than that of normal tissue [26-

32].

Many studies have shown a high Ki67 labeling index in various malignancies: brain

tumor [31], head and neck cancer [33,34], breast cancer [28,35-37], gastric cancer [30,38-42],

bladder cancer [27,43], rectal cancer [32] and prostatic cancer [44]. Ki67 (MIB-1) labeling

index is regarded as one of the indexes that show degree of malignancy of tumor in practical

work. In general, it is thought that a tumor that shows a higher level of proliferation activityhas a higher degree of malignancy.

Past studies have shown relationships between high Ki67 labeling index and poor

prognosis in breast cancer [28,35-37,45-47], bladder cancer [27,43], brain tumor [31], and

prostatic cancer [44]. However, other studies have shown that there are no relationships

between high Ki67 labeling index and poor prognosis in head and neck cancer [34], rectal

cancer [32], and gastric cancer [40-42]. A high level of cell proliferation activity of a

malignant tumor does not therefore appear to be simply correlated with poor prognosis of

patients. It appears that there is no correlation between high Ki67 labeling index and poor

prognosis for malignancies that have arisen from oral-digestive system (gastric cancer, rectalcancer, head and neck tumor, etc.), and there might be something like organ dependence in

the relationships between high proliferation activity level and prognosis. In the relationships


32/320


between other factors, histological grade [27,47], depth of tumor invasion [43], tumor size

[38], blood vessel invasion [41], expression of p53 [29], and lymphogenous metastases

[33,35,36] have been shown to be factors that correlate with Ki67 labeling index, but a

definite theory has not yet been established.

On the other hand, from the viewpoint of conventional radiation biology, it is known that

the higher the level cell proliferation activity becomes, the more radiosensitivity increases.

Based on this theory, a tumor with a high level of proliferation activity seems to be

radiosensitive but to have a higher degree of malignancy. It has not been determined whether

a high level of proliferation activity of a tumor is an advantage or disadvantage for patients

with esophageal carcinoma.

In our study, Ki67 labeling index was evaluated using sections of macroscopically

localized type and infiltrative type of esophageal carcinoma without preoperative treatment.

Ki67 labeling index was estimated independently by two skilled pathologists who had

received no clinical information other than the name of the disease, and several microscopic

fields were selected at random and the Ki67-positive cell rate (Ki67 labeling index) wascalculated by counting 1000 malignant cells. The average value of the two Ki67 labeling

indexes was taken as the Ki67 labeling index of the specimen (There was good agreement

between the Ki67 labeling indexes calculated by the two pathologists: mean ±S.D. values of

the two Ki67 labeling indexes were 53.3 ±20.7% and 55.1 ±21.3%, respectively (p=N.S.) and

the mean difference between two Ki67 indexes of the same specimen was 8.4 ±5.9%.).

The mean (±S.D.) Ki67 labeling index of all specimens was 54.2% (±20.4), and there

was a large variation (range, 8.5-88.5%). In comparison according to macroscopic type, Ki67

labeling index of the infiltrative type was significantly lower than that of the localized type

(46.9 ±20.4% vs. 61.5 ±18.1%, p=0.022, Figure 2A). The values of Ki67 labeling indexshowed a large variation from 8.5% to 88.5% in this study, and the values showed an almost

normal distribution.

Figure 2. Comparison between localized type and infiltrative type in Ki67 labeling index (A), average-

MVD (B), highest-MVD (C) and VEGF expression (D). Black circles and black bars: localized type,

white circles and white bars: infiltrative type, Ki67: Ki67 labeling index, average-MVD: average-

microvascular density, highest-MVD: highest-microvascular density, VEGF: vascular endothelialgrowth factor. Average-MVD: sum of vessel counts in four randomly selected fields in the tumor tissue


33/320


at low magnification. Highest-MVD: microvessel count at high magnification in the area of highest

neovascularization in the tumor.

According to past studies on Ki67 positivity, the ranges of Ki67 positivity were 11-18%

in brain tumor [31], 14-64% in esophageal cancer [48], 4-87% in gastric cancer [41], 7-70%

in colorectal cancer [32], 10-50% in cervical cancer [49], 0-17% in bladder cancer [27,43],

and 0-17% in prostatic cancer [44]. A certain variation of cell proliferation activity is seen in

almost all malignancies, but a very large variation is seen in malignancies of the digestive

system. The distribution of Ki67 labeling indexes in this study nearly corresponded with

these in past studies, but the reason for these variations is unknown. It seems to be important

that there was a significant difference between cell proliferation activities in macroscopically

localized type and infiltrative type of esophageal carcinoma. A high Ki67 labeling index was

expected in infiltrative type, which has an unfavorable prognosis, but contrary to that

expectation, the Ki67 labeling index of the infiltrative type was significantly lower than that

of the localized type.

Several studies have also suggested that there is no relationship between high Ki67

labeling index and poor prognosis in esophageal carcinoma [50-52]. It has been reported that

the prognosis of patients with a high Ki67 labeling index was more favorable than that of

patients with a low Ki67 labeling index who received radiotherapy for esophageal carcinoma

[51]. However, there was a large overlap of Ki67 labeling index between localized type and

infiltrative type of esophageal carcinoma in this study, and it cannot be concluded that

patients with a high Ki67 labeling index have good prognosis. The mechanism and cause of

the difference in cell proliferation activity is discussed in the following section with the

results of other factors.

4. ANGIOGENESIS AND PIVOTAL R OLE OF VEGF

Tumor cells keep growing while they are within a microscopic size, but the tumor

requires oxygen and nutrition when it grows larger than a certain size. Growth of a tumor in

diameter temporarily stops when it has reached a certain size, and then angiogenesis precedes

the next tumor growth in diameter [53]. Several conditions are required for the process of

angiogenesis, including demand for oxygen by the tumor, release of angiogenic factors from

tumor cells, attenuation of angiogenesis-inhibiting factors by the host, and migration of

endothelial cells. These mechanisms consist of interaction between tumor cells, host tissue,

and endothelial cells, and then formed blood vessels accelerate further growth of the tumor

[54-56].

Various angiogenic factors, including bFGF (basic fibroblast growth factor), PD-ECGF

(platelet-derived endothelial cell growth factor) and PlGF (placenta growth factor) have been

identified, but VEGF (vascular endothelial growth factor) is considered to be one of the

strongest angiogenic factors [57]. VEGF is a 34-42-kDa heparin-binding, dimeric, disulfide-

bonded glycoprotein. VEGF is known as VPF (vascular permeability factor), and it has been

shown that VEGF is a potent mediator of angiogenesis and vascular permeability [58-60].Expression of VPF/VEGF is seen in various animals, various normal organs, and various


34/320


neoplasms. VEGF stimulates secretion of fibrinogen to the extra-vascular matrix and

contributes to the formation of an interstitial fibrin structure of the tumor. Activities of the

fibrinolytic system and plasminogen affect interstitial fibrin formation, and the activity differs

depending on the tumors. Development of tumor stroma activates inflammatory reaction and

migration of macrophages [61,62]. There is something in common between tumor stroma

generation, tumor neovascularization, and the process of wound healing [63].

bFGF is thought to be one of the potent angiogenic factors, but VEGF, unlike bFGF,

specifically interacts with endothelial cells. It is known that VEGF, unlike the angiogenic

factor PD-ECGF, stimulates not only migration of endothelial cells but also proliferation of

endothelial cells [64,65]. Several studies in which the expressions of VEGF and bFGF were

compared showed that VEGF is more inducible and highly expressed by hypoxia than is

bFGF in experiments using xenografts of melanoma and pancreatic tumor cell lines [66-68].

Other experiments showed that a monoclonal antibody specific for VEGF, the inactivated

recombinant soluble human VEGF receptor, and a dominant-negative mutant of the VEGF

receptor inhibited angiogenesis of a tumor [69-71]. These findings suggest that VEGF playsan important role in angiogenesis, although there are many angiogenic factors, including

bFGF, Ang1/2 and PD-ECGF [72].

5. MICROVASCULAR DENSITY, OXYGEN TENSION, AND

HYPOXIC FRACTION

Tissue oxygen tension affects many intracellular environments such as glucose

metabolism, cell cycle, and angiogenesis. Oxygen tension is one of the important parameters

in tumor tissue, and various studies have been carried out to determine oxygen tension in a

tumor. Evaluation of microvascular density by immunohistochemistry [73-76], evaluation of

hypoxia by a hypoxia marker such as misonidazole and pimonidazole [66,77-80], evaluation

of blood perfusion by a perfusion marker Hoechst33342 [81], and measurement by

Eppendorf oxygen electrodes [75,82,83] have been performed for evaluation of tumor oxygen

tension. Oxygen tension in a tumor seems to be dependent on microvascular density and

blood vessel perfusion, and it has been shown that there is a significant correlation between

microvascular density and oxygen tension in human tumors [74,84].

It had been thought that cells within a certain distance from vessels were oxygenated andthat cells more distant from vessels were in a hypoxic state. However, not all hypoxic cells

exist in uniform distance from blood vessels because of heterogeneity in the actual tumor

[85,86]. Results of some studies have shown that there is a correlation between microvascular

density and tumor oxygen tension [75,76,82], while results of other studies have shown that

there is no correlation between microvascular density and tumor oxygen tension [87-89]. A

correlation between microvascular density and oxygen tension was seen in cervical cancer,

no correlation was seen in head and neck cancer, and both results that there were significant

correlation between microvascular density and oxygen tension and that there were no

significant correlation between them were observed in experiments using melanomaxenografts. These different findings suggest that there is heterogeneity depending on the

organ or cell line. However, biopsy specimens were used for evaluating microvascular


35/320


density in the above studies, and it is problematic whether a biopsy specimen displays

characteristics of the whole tumor. The uncertainty of these different results of the studies

may caused by methodological problems.

On the other hand, an experiment in which the proportion of hypoxic cells and distance

from blood vessels were determined using breast cancer xenografts showed that the

proportion of the radioresistant hypoxic fraction increased in an area more than 140 µm from

blood vessels [90]. To sum up, although there is individual heterogeneity due to oxygen

diffusion, blood perfusion, acidosis and glucose metabolism, it can be concluded that the

proportion of hypoxic cells increases with increase in intervascular distance in tumor tissue.

6. MICROVASCULAR DENSITY AND VEGF EXPRESSION

We discussed the concept of vessel density that affects tumor oxygenation and amount of

the hypoxic fraction of a tumor in the previous section. There seems to be another

significance in microvascular density. A local "vascular hot spot" induced by overexpression

of an angiogenic factor in tumor tissue seems to have another importance from the viewpoint

of tumor biology. Weidner et al. suggested this concept of "vascular hot spot" early on [73].

It is thought that a "vascular hot spot" indicates pathological angiogenic activity of the tumor.

The way to make vessel count to evaluate angiogenic activity is different from the way to

make vessel count to evaluate tumor oxygen tension. After the area of highest

neovascularization in a tumor specimen has been identified by observation under a

microscope at low magnification, vessel count is then made at high magnification in the

vascular hot spot. According to their study, a significant positive correlation between VEGFexpression and microvascular density (in the vascular hot spot) has been shown in human

breast cancer [91]. It is thought that microvessels in the vascular hot spot are induced by

VEGF overexpression. Correlations between VEGF expression and neovascularization have

also been found in breast cancer, hepatocellular carcinoma (HCC), ovarian cancer, germ cell

tumor, and melanoma xenografts [53,66,68,92-95].

In an experiment, the growth of VEGF-transfected xenografts was significantly faster

and the microvascular density of VEGF-transfected xenografts was significantly higher than

those of control xenografts [76]. Another study has suggested that genetic overexpression of

VEGF is more important than hypoxia-induced upregulation [66]. To sum up, microvasculardensity of a local "vascular hot spot" in tumor tissue significantly correlates with VEGF

expression and indicates activity of tumor angiogenesis. In analysis of tumor tissue obtained

from human non-small cell lung cancer (NSCLC), expression level of a hypoxia marker

(hypoxia-inducible factor: HIF family) was very high in specimens with high microvascular

density or low microvascular density but was low in specimens with medium microvascular

density [96]. These findings suggest that there are two patterns of vessel structure: one affects

oxygenation and formation of a hypoxic fraction of the tumor, and the other is hypoxia-

induced and angiogenic factor-activated microvessels.


36/320


7. R ESULTS OF OUR INVESTIGATIONS

We evaluated microvascular density of the tumor in two ways and expression of VEGF

by immunohistochemistry using surgically resected human esophageal squamous cell

carcinoma specimens. 1) Average microvascular density (a-MVD): a-MVD was estimated as

an index of tumor oxygen tension or amount of oxygenated cells. Using sections stained with

anti-CD34 antibody, four areas including tumor tissue were randomly selected, and

microvessel counts were made at low magnification, and then the sum of microvessels in the

four fields was taken as a-MVD of the section. 2) Highest microvascular density (h-MVD): h-

MVD was estimated as an index of the most active neovascularization in the tumor. Using

sections stained with anti-CD34 antibody, the area of highest neovascularization in the tumor

was identified and microvessel count of that field were made at high magnification.

Microvessels were counted on the basis of the methods described by Weidner [73]. 3) VEGF

expression: Several microscopic fields were selected at random, and VEGF-positive cell rate

was calculated by counting 1000 malignant cells.

In the analysis of average microvascular density, a-MVD of macroscopically infiltrative

type, which is thought to be radioresistant and have a poor prognosis, was significantly

higher than that of localized type (mean ±S.D.: 370 ±102 vs. 475 ±91, p=0.0014, Figure 2B),

[22]. In contrast, h-MVD of infiltrative type was significantly higher than that of localized

type (150 ±75 vs. 82 ±33, p=0.0006, Figure 2C). In the analysis of VEGF expression, VEGF

expression of infiltrative type was significantly higher than that of localized type (67.4 ±15

vs. 44.4 ±13, respectively, p


37/320


TP expression and highest-MVD. D) Survival curves according to VEGF expression. The patients were

divided into two groups of equal size according to VEGF expression. E) Survival curves according to

TP/VEGF expressions. TP(-)/VEGF(-): n=11, TP(-)/VEGF(+) or TP(+)/VEGF(-): n=18,

TP(+)/VEGF(+): n=11.

Different results have been obtained regarding the relationship between microvascular

density and tumor oxygen tension, the correlation between microvascular density and

hypoxic fraction, the relationship between microvascular density and radiosensitivity, and the

relationship between microvascular density and VEGF expression [66,74-76,80-84,87-89,92-

95,97]. One possible reason for the difference in results is the difference in methods used to

evaluate microvascular density. Another possible reason for heterogeneity of data is the

difference in type of study: an experimental study using an animal model, an in vitro study, or

a clinical study using human tumor. There seems to be a tendency that uniform data can

easily be obtained in an experimental study using cell lines with identical clones or using a

model of uniform animal xenografts, whereas it is difficult to obtain uniform data in a clinical

study using specimens from human tumors that consist of heterogeneous clones. Accuratedefinition in evaluating microvascular density should be required for avoiding confusion.

The results of our study showed paradoxical data that the inverse vessel counts were

obtained between average microvascular density and highest microvascular density in

comparison between macroscopically localized type and infiltrative type of esophageal

carcinoma. Based on known information, these data can be interpreted as follows. 1) The

number of hypoxic cells increases with increase in intervessel distance, and low average-

MVD in tumor tissue therefore leads to an increase in the amount of hypoxic fraction. Low

average-MVD in infiltrative type of esophageal carcinoma suggests low oxygen supply to the

tissue and the presence of a hypoxic fraction. The finding of a lower Ki67 labeling index(low cell proliferation activity) in the infiltrative type supports the presence of larger amount

of hypoxic fraction.

Figure 4. Presumptive mechanism of the formation of a hypoxic fraction in tumor tissue and

angiogenesis from the results of our study and known information. 1) There is underdeveloped tumor

vascularization (as suggested by low average-MVD). 2) The hypoxic fraction increases (as suggested


38/320


by low Ki67 labeling index). 3) VEGF is induced by hypoxia. 4) Development of immature

microvessels is induced by VEGF overexpression (as suggested by high highest-MVD). 5) These

microvessels with high permeability increase the incidence of hematogenous metastasis (which seems

to be one of the reasons for unfavorable prognosis).

2) Hypoxia-inducible genes and hypoxia-inducible proteins are expressed in the hypoxic

fraction. The potent angiogenic factor VEGF is one of the hypoxia-inducible proteins

[67,72,98,99]. The results showing low average-MVD and high VEGF expression level in

infiltrative type of esophageal carcinoma suggest the presence of hypoxia and hypoxia-

inducible VEGF expression (Figure 4A). It was not statistically strong, but negative

correlation between a-MVD and VEGF expression is thought to be one of the supporting

findings of relationship between increase in hypoxic fraction and increase in expression of

angiogenic factor.

3) Angiogenesis is activated by VEGF overexpression, and it was found that irregular

microvessels locally and densely developed in tumor tissue (Figure 4B). Low average-MVD,

high VEGF expression level, and high highest-MVD were seen in the infiltrative type ofesophageal carcinoma. The strong positive correlation between VEGF expression and

highest-MVD also seems to support the above mechanism. Figure 5(A-F) shows microscopic

findings of typical cases of macroscopically localized type (A-C) and infiltrative type (D-F).

The microscopic photos in A-C and those in D-F are of the same specimens: Figures A and D

show average-MVD (CD34 stain at low magnification), Figures B and E show VEGF

expression and Figures C and F show highest-MVD (CD34 stain at high magnification).

Figure 5. Microscopic findings of typical cases of macroscopically localized type (A-C) and infiltrative

type (D-F) of esophageal carcinoma. The microscopic photos in A-C and those in D-F are of the same

specimens. A/D: CD34 stain at low magnification (for estimating average-MVD), B/E: VEGF stain,

C/F: CD34 stain at high magnification (for estimating highest-MVD). The case of macroscopically

infiltrative type of esophageal carcinoma (D-F) shows low average-MVD (D), overexpression of VEGF(E) that seems to be induced by hypoxia, and activated neovascularization (F) that seems to be VEGF-

induced vascularization.


39/320


4) Furthermore, these immature vessels with high permeability stimulate the occurrence

of hematogenous or lymphogenous metastasis [91,92]. The presence of a hypoxic fraction not

only leads to radioresistance but also increases the frequency of hematogenous metastasis.

The mechanism such like this (abovementioned 1-4) seems to be one of the causes of

radioresistance and unfavorable prognosis in the infiltrative type of esophageal carcinoma.

There were large variations in the results of Ki67 labeling indexes and microvessel counts in

our studies. This is because the data were obtained from human specimens consisting of

heterogeneous clones and conditions. Although it is difficult to obtain uniform data in a study

using human tumor specimens consisting heterogeneous subjects, it is noteworthy that

significant differences were found despite the large variations.

8. HYPOXIC FRACTION AND CELL CYCLE

It has generally been thought that the amount of quiescent cells increases in the hypoxic

fraction [100,101]. The results of a study in which hypoxic fractions were compared using

melanoma xenografts showed there were relationships between low microvascular density,

increase in hypoxic fraction, and increase in the proportion of quiescent cells by flow

cytometry [97]. However, it has been reported that not all cells in the hypoxic fraction are in

quiescent phase and that some cells proliferate slowly even in the hypoxic area. Rate of

proliferating cells in the hypoxic fraction seems to differ depending on cell line, but it is

thought that malignant cells stop or delay their cell cycle at G0/G1 phase or at G2/M phase

[101,102].

The results of our study showed that the cell proliferation activity level of infiltrativetype of esophageal carcinoma, which seems to be radioresistant and have an unfavorable

prognosis, was significantly lower than that of localized type. This was unexpected because it

is generally thought that a tumor with a high growth rate is highly malignant. However, to

sum up following factors such as increase in hypoxic fraction, cell cycle stop/delay,

overexpressions of hypoxia-inducible proteins (VEGF) and being radioresistance, it is

consistent that infiltrative type of esophageal carcinoma that is thought to be radioresistant

and to be refractory to treatment could be poorly oxygenated and show low level of

proliferation activity.

A study has shown that mutation of p53 causes alteration in the cell cycle under acondition of hypoxia and that this leads to resistance to hypoxia [103]. It has also been

reported that normal endothelial cells in S phase increase under a condition of hypoxia and

that the cell cycle of normal endothelial cells is delayed but does not stop under a condition

of hypoxia [104]. This seems to be rational mechanism, but it is interesting that endothelial

cells, which play a pivotal role in angiogenesis, do not stop their cell cycle and continue to

proliferate even under a condition of hypoxia.


40/320


9. HYPOXIA-INDUCIBLE FACTORS (HIFS) ANDMETABOLIC

CHANGES

Metabolism under a condition of hypoxia that differs from metabolism under the

condition of normoxia. Under hypoxic conditions, an increase in blood perfusion is seen innormal tissue by its ability of autoregulation [105]. However, it is not known whether tumor

tissue has the ability to autoregulate blood perfusion. It is generally thought that angiogenesis

and oxygen supply cannot meet oxygen consumption by preceding tumor growth, and then

the hypoxic fraction in the tumor usually increases with tumor growth.

An in vivo experimental study showed that there is a gradient of oxygen tension in tumor

tissue and that inc

Documents

11460.Oral Cancer Research Advances by Alexios P. Nikolakakos