Download Problem Statement

Thesis Proposal
Presented to:
Dr. Vijay Vaidyanathan
Submitted by:
Mehul B.Baxi
1
Introduction
Every year 31,000 Americans are afflicted with oral cancer [1]. It is believed that
early diagnosis could significantly increase the survival rate from 50% to about 80% [2].
Fluorescence spectroscopic detection (FSD) and photodynamic therapy (PDT) may
provide an effective approach for early detection and treatment of oral cancer [2].
Autofluorescence, the natural fluorescence of tissue, has been used to distinguish
between normal and neoplastic tissue.
The differentiation is largely dependent on the biochemical composition and
histomorphologic structure of the tissues that undergo a change during transformation
from normal to malignant. More prominent differences were found at 410 nm [3].
Photosensitizers that are more selectively retained in neoplastic tissues could enhance the
reliability of optical diagnosis and photodynamic therapy. Fluorescent dyes such as
photofrin have been used in studies on photodynamic therapy. Photosensitizers like
photofrin have been widely used in photodynamic therapy and tissue differentiation;
however, adverse side effects such as prolonged cutaneous photosensitization have
limited their use. A higher porphyrin accumulation and PDT selectivity in the colon
carcinoma model due to ALA – induced PpIX than photofrin [4]. This shall be used for
monitoring of ALA induced PpIX fluorescence and autofluorescence, as a non-invasive
method of differentiating normal and cancerous tissue [5].
2
Problem Statement
Tissue differentiation by the optical non-invasive technique for detecting
differences in the cancerous and non-cancerous tissues. Mat Lab shall be used for
analyzing details and characteristics of different types of cancerous and non-cancerous
tissues. Also statistical analysis software shall be used for performing the statistical
analysis.
5-Aminolevulinic acid (ALA)–induced protoporphyrin IX (PpIX) is a promising
photosensitizer that could enhance the spectroscopic contrast between normal and
diseased oral tissues. Knowledge of the pharmacokinetics and effects on tissue type are
important for diagnostic and therapeutic procedures.
Earlier research conducted experiments and recorded the fluorescence from
buccal mucosa, gums, tongue, and facial skin using a fiberoptic probe connected to an
optical multichannel analyzer. Blood samples were collected for hematologic and serum
biochemical analysis. Pharmacokinetic parameters of interest were estimated using a
compartmental model. PpIX fluorescence at all sites reached a peak in 2–6 hours, and
returned to baseline in 24–31 hours, depending on the dose. Plasma PpIX peaked earlier
than oral tissues. The rate of synthesis of PpIX, and its conversion to heme products are
dose dependent. Different tissues have different pharmacokinetic response.
3
Background of the Study
ALA is a naturally occurring precursor of heme. The native compound is not a
photosensitizer, but in certain types of cells and tissues, it is metabolized to the
photosensitizer PpIX in the body [2]. Intracellular porphyrin localization and
photoactivation have been reported to occur first within the plasma membrane. After
several hours of incubation, redistribution within the cell occurs to include the nuclear
membrane and other organelles such as mitochondria and lysosomes [6, 7]. When excited
with suitable wavelength of light (~ 410 nm), PpIX emits characteristic fluorescence in
the red spectrum (636 nm). In a recent in vitro study the subcellular mitochondrial
localization of the photodynamic sensitizers photofrin and ALA-induced PpIX in
radiation-induced fibrosarcoma (RIF) tumor cells was investigated [8]. As with photofrin,
ALA-induced PpIX showed weaker localization in the mitochondria in RIF-8A than in
RIF-1 cells.
A hamster cheek-pouch model, experiment that was conducted reported a
significant increase in tissue autofluorescence and ALA-induced PpIX fluorescence
during transformation of buccal mucosa from normal to premalignant to malignant. The
fluorescence measurements were found to correlate well with histopathologic assessment
of tumor development. Another research showed that oral squamous cell carcinomas
could synthesize and accumulate photosensitizing levels of PpIX after oral ALA
administration. In recent studies, PpIX fluorescence was detected in the oral mucosa of
16 patients after topical application of ALA [9]. PpIX in neoplastic tissue was found to
accumulate earlier than in nonneoplastic tissue with maximum fluorescence contrast seen
1–2 hours after application [9]. The efficacy of photodynamic therapy by using ALA for
4
premalignant and malignant lesions of the oral cavity after oral administration of a dose
of 60 mg/kg and treatment with laser light at 628 nm (100 J/cm2). The depth of necrosis
was found to vary from 0.1 to 0.3 mm, but complete epithelial necrosis was found to be
present in all cases [9].
ALA-induced PpIX is a promising photosensitizer that could be used in
noninvasive diagnosis and photodynamic therapy of oral cancers [10]. Because the oral
cavity is easily accessible with flexible optical fibers, it is a suitable site for fluorescence
detection. Our study has particular relevance to the use of ALA for diagnosis and therapy
of oral cancers in veterinary medicine.
The aim is to investigate the detection of non-cancerous and cancerous tissue
depending on the dosage and the ALA-induced fluorescence kinetics. Because anesthesia
may affect the animal’s metabolism and, thus, the production of PpIX at the site of
interest, awake dogs will be studied. Although there have been fluorescence studies
conducted with ALA on small animals such as hamsters, there are few studies reported
on larger animals. Before use in humans, clear understanding of the effect of dosage and
tissue type on ALA induced fluorescence is needed. Dogs, like humans, are known to
develop oral cancer naturally, and their oral cavity is similar to that of humans. Thus,
dogs are a good model in which to study the pharmacokinetics of ALA.
5
Purpose of the Research/ Study
The purpose of this research is application of advanced technology in the
detection of cancer research. A compact fluorescence spectroscopic tool for in vivo point
monitoring of ALA - induced PpIX fluorescence and autofluorescence, as a non-invasive
method of differentiating normal and cancerous tissue. Mat Lab will be used here to
assist in the process of tissue differentiation.
The main work points in this study are:
1. 4-5 dogs selected and used for the study.
2. Tissue auto fluorescence to be measured with the optical fiber probe using an
optical multi channel analyzer (OMA).
3. Determine non-invasive techniques for tissue differentiation between the
cancerous tissues and normal tissues.
6
Research Questions
This thesis will attempt to provide a detailed knowledge of the optical noninvasive technique used for the detection of oral cancer in dogs. It shall attempt to answer
the following questions:
1. How does optical non-invasive technique benefit in the detection of oral cancer?
2. Is there any relation between the waveforms obtained for the intensity of the
fluorescence and the wavelength of light?
3. Is there a mathematical relation between the cancerous tissues after using
statistical techniques for performing discriminate analysis?
4. What significant capabilities can be accomplished if the detection of oral cancer
is successfully shown by the calculations done on the fluorescence waveforms?
7
Review of Literature
From the review of literature it has been found that new techniques for noninvasive early detection and diagnosis of oral dysplasia and carcinoma are required [11].
Fluorescence spectroscopic detection (FSD) and photodynamic therapy (PDT) may
provide an effective approach for early detection and treatment of oral cancer.
Autofluorescence, the natural fluorescence of tissue, has been used to distinguish
between normal and neoplastic tissue [10]. The use of autofluorescence spectra at
excitation wavelengths of 370 and 410 nm to distinguish between normal oral mucosa
and abnormal tissue samples and found more prominent differences at 410 nm [3].
The system to be used for fluorescence monitoring can be used in a variety of
clinical applications [4]. Light-induced fluorescence after ALA exposure can differentiate
between the different stages of premalignancy and malignancy. Its ability to differentiate
between healthy tissue and early pathology is particularly interesting [11]. This could be
done using a compact fluorescence spectroscopic tool for in vivo point monitoring of
aminolaevulinic acid (ALA)-induced protoporphyrin IX (PpIX) fluorescence and
autofluorescence, as a non-invasive method of differentiating normal and cancerous
tissue [5].
A low-cost system for the illumination can also be used [12]. Such a developed
system shows the capability of differentiating between different histopathological stages
of oral lesions, suggesting a significant potential for realizing the non-invasive optical
biopsy for early cancer diagnosis [13]. As a fluorescent marker, PPIX could represent a
8
possible new diagnostic tool to detect early malignant and secondary lesions in the oral
cavity. In addition, 5-ALA-induced PPIX fluorescence is promising as a useful
intraoperative tool for determining adequate surgical margins of resection. Further
investigations aim to assess this diagnostic procedure as a sensitive and clinically reliable
method for patients with oral cancer [14].
A quantitative analysis of the fluorescence contrast in neoplastic and surrounding
tissue was performed using an optical multichannel analyzer (OMA) [15]. Further
investigations are required to assess the value of this new diagnostic procedure as a noninvasive and sensitive method for patients with head and neck cancer not only in pre- and
postoperative diagnostic studies but also for a fluorescence-guided resection of tumors
[15].
9
Materials and Methods
The study is experimental. Cancerous dogs of mixed breed will be used in this
study. Three target sites are selected in the oral cavity for the fluorescence detection.
These target sites are the buccal mucosa, tongue and the gums. Tissue auto fluorescence
will be measured from all the three sites of interest.
The light source will be a 150W xenon lamp, filtered to provide ultraviolet light
for excitation (400 ± 30nm). The laser will be tuned to 410 nm. A sheathed bifurcated
fiber optic probe will be used to provide excitation and collect fluorescence. This
fluorescence measurement shall be done in the dark.
Fig: Set Up of the Experiment
The dogs shall be fasted overnight. The dogs will be given an intra muscular
injection of a pre anesthetic acetyl promazine (0.11 mg/kg). A catheter is inserted in the
cephalic vein.
10
5-ALA will be dissolved in sterile phosphate buffered saline at a concentration of
80 mg/ml and immediately administered through the cephalic catheter over 5 min. The
ALA induced fluorescence will be measured from our target sites.
All the fluorescence data shall be recorded as the ratio of intensities observed at
wavelength of red light that is at 636 nm and wavelength of orange light that is 595 nm.
The red wavelength band is significant because of the characteristic PpIX fluorescence
peak in the red spectrum, whereas the orange wavelength is chosen to minimize the effect
of tissue optics on measurement.
Dose of ALA of 25 mg/kg will be administered. A wash-out period of 1 week
between doses will ensure that residual PpIX will not be present. The fluorescence data
shall be collected from the target areas at 1, 2, 3, 5, 7, 14, 24 hours respectively after the
administration of ALA. At each site 3 values will be taken, and their values averaged.
The dogs shall be housed in a darkened kennel for 24 hours after treatment to prevent
cutaneous photo toxicity.
A classification model shall be determined from the spectra that are obtained from
the tissue readings. The lineshape of each fluorescence spectrum will be determined by
dividing the spectrum from 430 nm to 700 nm. Each fluorescence lineshape spectrum
will be sampled at 15 nm intervals from 430 nm to 700 nm. This will result in discrete
points that shall be used for the statistical analysis. Because tissue fluorescence emissions
have a broadband spectral shape without any sharp features, a 15 nm sampling will
represent the system without any loss of information [16].
The intensities at the sampled wavelengths will be used as the independent
variables in a stepwise discriminate analysis. The dependent variable (Intensity) has two
11
non-numeric values N for normal and C for cancer. Using the stepwise discriminate
analysis, the significant wavelengths were selected.
The intensities at the selected wavelengths will be used in formulating a
classification model using linear discriminate analysis. After determination of the
classification model, the fluorescence lineshape will be used to determine the sensitivity
and specificity of the model in classifying malignant and normal tissues.
FFT shall be calculated for the given signals.
The figures below show the FFT waveforms of a dog Honey 1 hour after the
application of ALA in a normal as well as a diseased dog. This is the analysis done on the
previous dog from an earlier research.
Fig: FFT performed 1 hour after application of ALA for a Normal Dog
12
Fig: FFT performed 1 hour after application of ALA for a Diseased Dog
From the above waveforms a characteristic feature that can be seen is the
presence of peaks in the phase plot. At this stage some sort of relationship is being
worked upon to try to establish between the plots of the normal and diseased dogs. This
would be quite similar to the current research where a diseased dog would be analyzed
for such details.
A conclusion shall be drawn at the end after studying the classification model and
after performing the FFT.
13
Limitations
This research discusses only limited applications of the method to distinguish
between normal and neoplastic tissue. The research will not provide a detailed
description of all types of tissue differentiation techniques possible under this study. It is
not claimed that this is the best and the only technique for tissue differentiation available.
There might be other techniques that might be more efficient than the one used.
One of the main limitations in this research is the availability of the dogs and the
OMA at the same time.
Another factor that would be a constraint is the following of time and to complete
the experiments and data analysis by the end of September 2003.
14
Summary of the Chapter
Fluorescence spectroscopy, which has emerged as a potential non-invasive
diagnostic tool to detect pre malignant changes in oral tissues. Thus there is a need for a
photosensitizer that could enhance the spectroscopic contrast between normal and
neoplastic tissue, while allowing for selective photosensitization of pre malignant and
malignant lesions in the oral cavity. It is proposed to extend earlier findings, initiate
research on dogs with naturally occurring neoplasia, and determine if differentiation
between neoplastic and normal tissue can be accurately determined using fluorescence
spectroscopy.
15
Time Line
The research has been started in Spring 2003 and will be completed by Dec 2003.
If necessary due to some unexpected situation, it might extend to Spring 2004. A
proposed rough time line is: Review of literature and proposal approval should be
completed by end of March. Conduct experiments and mathematical analysis of the
fluorescence waveforms and other related results, which should complete in Summer and
Fall of academic year 2003. Completion of thesis, a manuscript of a paper for journal
shall be done by the end of the year 2003. A more detailed time line will be provided if
the study is approved.
16
Recommendation for further study
Present research is conducted to determine a method that can accurately identify
margins between the neoplastic and the normal tissue by performing FFT on the intensity
– wavelength data. Once this method is ascertained and proved it should be modified and
applied for doing the following:
1. Check out the detection for oral cancer in humans using the current method used
in this study.
2. Find out a method that would be even simpler and productive than the current
method.
17
Hypothesis
Null Hypothesis
There is no difference between optical non-invasive technique for tissue
differentiation as compared to the regular method of observation before performing the
biopsy.
Alternative Hypothesis
There is difference between optical non-invasive technique for tissue
differentiation as compared to the regular method of observation before performing the
biopsy.
18
Expected Outcomes
Once the FFT has been calculated, appropriate discriminate analysis will be
performed which can accurately identify margins between the neoplastic tissue and the
normal tissue. This will make the procedure of detection a lot easier than going through
the regular method of performing a biopsy and by microscopic observation.
19
Significance/ Scope of the Study
New techniques for non-invasive early detection and diagnosis of oral cancer are
the need of the medical field today. Research in this field for the detection of oral cancer
is very advantageous because dogs, like humans, are known to develop oral cancer
naturally, and their oral cavity is similar to that of humans. Thus, dogs are a good model
in which to study the pharmacokinetics of ALA. Also the oral cavity is easily accessed
with a fiber optic probe; hence, it is suitable site for the fluorescence diagnostics.
Fluorescence ratio measurements made after the administration of ALA could assist in
the differentiation of neoplastic tissue from normal tissue, especially during the early
stages of neoplasia.
20
Bibliography
1.
Silverman S, and Gorsky M, "Epidemiologic and demographic update in oral
cancer: California and national data- 1973 to 1985," JADA, 120, p 495-499, 1990.
2.
Kennedy JC, Pottier RH, "Endogenous protoporphyrin IX, a clinically useful
photosensitizer for photodynamic therapy," Photochem Photobiol, 14, p 275–292,
1992
3.
Ingrams DR, Dhingra JK, Roy K, Perrault DF, Bottrill ID, Rebeiz EE, Kabani S,
Pankratov MM, Manoharan R, Itzkan I, Shapshay SM, Feld MS,
"Autofluorescence characteristics of oral mucosa," Head Neck, 19, p 27–32, 1997.
4.
Orenstein A, Kostenich G, Roitman L, Schectman Y, Kopolovic Y, Ehrenberg B,
Malik Z, "A comparative study of tissue distribution and photodynamic therapy
selectivity of chlorin e6, photofrin II and ALA-induced protoporphyrin IX in a
colon carcinoma model," Br J Cancer, 73, p 937–944, 1996.
5.
Nadeau Valerie, Hamdan Khaled, Hewett Jacqueline, Makaryceva Julija, Tait
Iain, Cushieri Alfred, Padgett Miles, "A compact fluorescence spectroscopic tool
for cancer detection," Proceedings of SPIE - The International Society for Optical
Engineering, v 4613, p 35-40, 2002.
21
6.
Berns BW, Dahlman A, Johnson FM, Burns R, Sperling D, Guiltinan M, Siemens
A, Walter R, Wright W, Hammer-Wilson M, Wile A, "In vitro cellular effects of
hematoporphyrin derivative," Cancer Res, 42, p 2325–2329, 1982.
7.
Shulok JR, Wade MH, Lin CW, "Subcellular localization of hematoporphyrin
derivative in bladder tumor cells in culture," Photochem Photobiol, 51, p 451–
457, 1990.
8.
Wilson BC, Olivo M, Singh G, "Subcellular localization of photofrin and
aminolevulinic acid and photodynamic cross-resistance in vitro in radiation
induced fibrosarcoma cells sensitive or resistant to photofrin-mediated
photodynamic therapy," Photochem Photobiol, 65, p 166–176, 1997.
9.
Leunig A, Rick K, Strepp H, Gutmann R, Goetz AE, Baumgartner R, Feyh J,
"Fluorescence imaging and spectroscopy of 5-aminolevulinic acid induced
protoporphyrin IX for the detection of neoplastic lesions in the oral cavity," Am J
Surg, 172, p 674–677, 1996.
10.
Vaidyanathan V.,Rastegar Sohi, Fossum T.W., Flores P., Breggen E.W., Egger
N.G., Jacques Steven L., Motamedi Massoud, "In-vivo kinetics of ALA-induced
fluorescence in the canine oral cavity," Proceedings of SPIE - The International
Society for Optical Engineering, v 2975, p 222-226, 1997.
22
11.
Ebihara A, Krasieva TB, Liaw LH, Fago S, Messadi D, Osann K, Wilder-Smith
P., "Detection and diagnosis of oral cancer by light-induced fluorescence," Lasers
Surg Med, 32(1), p 17-24, 2003.
12.
McKechnie Tracy, Hewett Jacqueline, Sibbett Wilson, Padgett Miles, Clark
Colin, Ferguson James, "Optimization of 5-aminolaevulinic acid application and
light excitation to ensure maximum contrast between cancerous and healthy tissue
on the skin," Proceedings of SPIE - The International Society for Optical
Engineering, v 3592, p 20-26, 1999.
13.
Wei Zheng, Olivo M., Sivanandan R., Karuman P., Tuan Kay Lim, Soo Khee
Chee, "Endoscopy imaging of 5-ALA-induced PPIX fluorescence for detecting
early neoplasms in oral cavity," Proceedings of the SPIE - The International
Society for Optical Engineering, v 4597.
14.
Leunig A, Betz CS, Mehlmann M, Stepp H, Arbogast S, Grevers G, Baumgartner
R., "Detection of squamous cell carcinoma of the oral cavity by imaging 5aminolevulinic acid-induced protoporphyrin IX fluorescence," Laryngoscope,
110(1), p 78-83, 2000 Jan.
15.
Leunig A, Rick K, Stepp H, Goetz A, Baumgartner R, Feyh J, "Photodynamic
diagnosis of neoplasms of the mouth cavity after local administration of 5aminolevulinic acid," Laryngorhinootologie, 75(8), p 459-64, 1996 Aug.
23
16.
Panjehpour M, Overholt BF, Schmidhammer JL, Farris C, Buckley PF, Vo-Dinh
T, “Spectroscopic diagnosis of esophageal cancer: new classification model,
improved measurement system,” Gastrointest Endosc, 41(6), p 577-81, 1995.
Assumptions
Definition of Terms
24