Blood content and tumor oxygen level are important biomarkers and prognostic indicators in patients with colorectal cancer (CRC). However, noninvasive measurements of both quantities in human colon are limited. In this study, we extracted the total hemoglobin concentration (THC) and oxygen saturation (StO2) of normal, premalignant, and malignant colonic tissues in 27 patients using a diffuse reflectance instrument and algorithms based on the diffusion equation. The mean±standard error of THC and StO2 from all normal sites (n=26) is 93.4±17.1μM and 67.2±3.7%, respectively. THC increased to 136.9±23.8μM and 153.8±38.6μM and StO2 decreased to 51.3±7.0% and 26.4±6.1% for premalignant and malignant tissues, respectively. The disease-to-normal THC ratios are 3.2±1.1 and 4.4±1.9 and the disease-to-normal StO2 ratios are 0.7±0.1 and 0.5±0.1 for premalignant and malignant tissues, respectively. These results demonstrate the feasibility of a robust optical method to assess colon THC and StO2 at all stages of carcinogenesis in vivo so that the angiogenesis and hypoxia of the disease and the therapeutic role can be studied in CRC patients.
© 2009 Optical Society of America
Angiogenesis is a hallmark of human malignancies . During normal organogenesis, the growth of new blood vessels (i.e. the process of angiogenesis) is highly regulated and balanced between angiogenesis inducers and inhibitors. Tumors appear to activate an “angiogenic switch” by favoring the changing of angiogenesis inducers than inhibitors [1, 2]. To determine tumor angiogenesis in the clinic, the ex vivo microvessel density of biopsy specimens is used. In vivo methods of quantitating angiogenesis adapt in vivo imaging techniques with the contrast agent or angiogenesis marker. Magnetic resonance imaging (MRI) of blood volume and vessel size index through an iron-based MR contrast agent has shown a good correlation with the ex vivo histological microvessel density [3, 4]. Vascular endothelial growth factor (VEGF), a key angiogenesis inducer, is the most commonly used and a specific molecular marker of angiogenesis. Many tumors including colorectal cancer (CRC) evidence increased VEGF expression associated with disease progression and poor prognosis [5, 6]. Recently, anti-VEGF therapeutics have entered the clinic and shown to improve survival in patients with metastatic CRC . In vivo imaging VEGF is achieved by MRI, single-photon emission computed tomography, positron emission tomography, computed tomography, ultrasound, and optical imaging through either non-radionuclide or radionuclide labeling .
Hypoxia develops within solid tumors and is found to induce malignant progression, increase metastatic potential, and increase resistance to all cancer treatments . In vivo technique based on the Eppendorf pO2 electrode is the current “gold standard” of clinical assessment of tumor oxygen level [9, 10]. Due to the invasiveness and lack of universal applicability of the pO2 electrode, ex vivo techniques based on exogenous or endogenous hypoxia markers become the alternative in the clinical setting. Hypoxia marker, 2-(2-nitroimida-zol-1[H]-yl)-N-(2,2,3,3,3-pentafluoropropyl)accetamide (EF5), has been used in human brain tumors, intraperitoneal carcinomatosis and sarcomatosis [11, 12], and shown to associate with tissue oxygen (pO2) . Hypoxia inducible factor-1 (HIF-1) is considered as an intrinsic marker of tumor hypoxia and up-regulates angiogenesis. However unlike VEGF in angiogenesis, HIF-1 may not be a specific hypoxia marker because HIF-1 has been shown to have no or weak correlation with pO2 [10, 14] and have no prognostic significance in the cervix cancer [10, 14] and CRC . Nevertheless, hypoxia is tightly related to angiogenesis because under hypoxia stress both HIF-dependent and HIF-independent (e.g. oncogenic Ras-medicated pathways) pathways control angiogenesis .
The intrinsic absorption spectral fingerprint of oxyhemogloblin (HbO2) and deoxyhemoglobin (Hb) at the visible to near infrared wavelength range allows optical methods to quantitate tissue angiogenesis and hypoxia noninvasively. Once the concentration of HbO2 (cHbO2) and Hb (cHb) is known, the total hemoglobin concentration (THC) and tissue blood oxygen saturation (StO2) can be calculated by summing cHbO2 and cHb (i.e. THC=cHbO2+cHb) and by dividing cHbO2 by THC (i.e. StO2= cHbO2/THC), respectively. Diffuse optical spectroscopy and imaging based on the diffusion equation (P1 approximation), for example, have been used noninvasively to measure human tissue THC and StO2 in vivo in breast , brain , intraperitoneal tissues , muscle , and head and neck . In the measurement of colonic tissues, Zonios, et al. have used a modified diffusion equation to model the reflectance from single source-detector separation (<1 mm) and derive the THC and StO2 of normal mucosa and adenomatous polyps on 13 patients undergoing routine colonoscopy .
The purpose of this pilot study is to establish a robust optical method to assess THC and StO2 of colonic tissues at all stages of colon carcinogenesis in vivo so that the angiogenesis and hypoxia of the disease and its therapeutic role can be studied in CRC patients. We achieve this aim by quantifying the THC and StO2 of normal, premalignant, and malignant colonic tissues of two normal, 17 polyp, and eight cancer patients undergoing routine colonoscopy using diffuse reflectance spectroscopy (DRS). Different from Zonios, et al. , a wavelength range of ~600 to 800 nm and two to three source-detector separations (~0.6 to 2.5 mm) were used in our study. We modified our previously developed algorithm based on the diffusion approximation [19, 23] to include a hybrid P3 diffuse reflectance model developed by Hull and Foster  in carrying out the reflectance obtained at a shorter separation (0.6 mm) and highly absorbing samples (e.g. increased hemoglobin concentration in tumors due to angiogenesis) where the diffusion approximation may be questionable. The choice of the range of the separations is to cover a wider range of colon wall that was reported to vary between ~1 mm to 3 to 5 mm in normal sites [25, 26] and became thicker in abnormal sites . We evaluated our P1 and hybrid P3 algorithms using tissue-simulating phantoms containing intact human red blood cells before analyzed the clinical data. Similar results to Zonios, et al. were found in our study that adenomatous polyps have higher THC than normal colon. Additionally, we have observed increased THC, as well as decreased oxygenation, during carcinogenesis. Our investigation takes the first step toward further studying the role of angiogenesis and hypoxia in the disease progress and prognosis of CRC patients noninvasively. Finally, we also compared the performance of our algorithms based on the P1 approximation and hybrid P3 model in tissue-simulating phantoms at studied wavelength ranges and source-detector separations.
2. Materials and methods
2.1 Human subjects and experiment methods
Noninvasive in vivo DRS measurements were collected from a total of 27 patients undergoing routine colonoscopy. Two, 17, and eight patients were diagnosed with normal, adenomatous polyps, and adenocarcinoma, respectively. All patients signed a study-specific informed consent and all experimental procedures were approved by the Institutional Review Board of the Taipei Veterans General Hospital. DRS measurements, through an optical fiber probe which was advanced through the accessory channel of the colonoscope and brought into contact with the tissue, were performed on diseased sites (i.e. adenomatous polyps or malignant tumors) and on the corresponding adjacent normal sites from all patients except normal patients where only measurements of normal sites were taken. Between one and four DRS measurements were taken for averaging from each tissue type in each patient; no more measurement could be taken due to time constraints during colonoscopy. The data acquisition time was 100 to 300 ms per measurement to maximize the intensity of collected reflectance without saturating the detector. During DRS measurements, the colonoscopic illumination light was turned off.
The polyp and tumor tissue samples were removed as the routine colonoscopy procedure for pathology examination immediately after in vivo DRS measurements were obtained. The size of the polyps ranges from 0.2 to 2 cm in diameter based on the endoscopic estimate and pathological report. The tumors were removed subsequently in the operative room. The size and pathology report of the tumor were then obtained.
2.2 Diffuse reflectance spectroscopy (DRS) system and analytical model
A continuous wave, white light diffuse reflectance spectroscopy (DRS) system was built based on previous published studies [19, 23] to collect reflectance spectra from tissue surface at multiple source-detector separations. Briefly, this DRS system was composed of a 100 W quartz tungsten halogen lamp (Olympus, Japan), a hand-held surface contact fiber-optic probe (Fiberoptic Systems Inc., Simi Valley, California), a spectrograph (SP-2156, Princeton Instruments-Acton, Acton, Massachusetts), and a -70 °C thermoelectrically cooled CCD camera (PIXIS:400BR, Princeton Instruments Inc., Trenton, New Jersey). The fiber-optic probe is consisted of a single source and 4 detection fibers (each with a core diameter of 200 um, numerical aperture 0.22). These fibers were packed together side-by-side as shown in Fig. 1(a) with source-detector separations, ρ, less than 5 mm (ρ = 0.6, 1.5, 2.5, and 4 mm). Each fiber tip was cut to 45° and polished to turn the delivered or collected light at 90°. All fibers were packed inside a Teflon tube with an outer diameter ~1.6 mm (Fig. 1(a)) so that the probe is thin enough to be inserted through the accessory channel of the colonoscope and then directly contact with colonic tissues (Figs. 1(b) and 1(c)) for optical measurements. As the previously published method [19, 23], the spectra acquired by this system were corrected for the dark count and the systematic spectral features of the source and detection system during a ‘calibration’ run using an integrating sphere (Labsphere, North Sutton, New Hampshire) to become Rmeasure. Then a source-detector-separation normalized reflectance, Rnorm, was calculated by dividing Rmeasure taken at any separation ρ by that taken at a fixed separation distance ρ0, typically the minimum separation, i.e. R(ρ,λ,)norm ≡ Rmeasured(ρ, λ)/ Rmeasured(ρ0, λ). In fitting the data we computed a similar Rnorm using calculated reflectance spectra Rcalculated(ρ, λ) by employing the analytical solution of the diffusion equation or a diffusion hybrid P3 model with semi-infinite boundary conditions to simultaneously fit all data in the optimal wavelength range and using optimal source-detector separations. In order to be consistent for all tissue types, only signals at source-detection separations at 0.6, 1.5, and 2.5 mm were used. The detected tissue depth is approximately 1/3 to 1/2 of the source-detection separations based on a Monte Carlo simulation run (data not shown) and thus is ~0.8 to 1.2 mm in this study. In some normal tissue sites, only separations at 0.6 and 1.5 mm were used (i.e. detection depth ~0.5 to 0.75 mm below the tissue surface) because of poor signals at larger separation that is likely due to the limited thickness of normal colon wall. The choice of the wavelength range was to minimize a fitting error and stabilize the derived parameters as described previously .
The hybrid P3 model is one of several implementations of a P3 approximation developed by Hull and Foster  that has computational simplicity by adapting the expression of the diffuse reflectance but provides accuracy comparable with other more complicated P3 approximations. It has been applied in monitoring carbogen-induced oxygenation changes of murine tumors in vivo . In the diffusion approximation with the extrapolated boundary condition, the diffuse reflectance R from homogeneous semi-infinite media is equivalent to the total reflectance from a pair of isotropic point sources in an infinite medium as show in Eq. (1) [19, 24]. The source is positioned at z0 underneath the physical semi-infinite surface and the image source is positioned at 2zb+z0 above the physical surface. The light distribution resulting from fiber-based illumination of a semi-infinite sample has been considered and modeled using extrapolated boundary conditions .
Here D is the diffusion constant, given by[3(μa + μs′)-1 [19, 24]. zb is the distance of the extrapolated boundary above the physical semi-infinite boundary equal to 2×2.95×D. z 0 = (μa + μs′)-1 and μeff is the effective attenuation coefficient defined by . r is the distance between either source or image source and the detection point that . ρ is the physical source-detector separation as defined earlier. c 1 and c 2 account for explicitly in the relative weights of the flux and fluence terms in the expression for reflectance , and are 0.1178 and 0.3056, respectively. These numbers are determined by the relative indices of tissue and detection fibers and by the numerical aperture of the detection fibers, which in our case is 0.22.
In hybrid P3 approximation, the reflectance is contributed by two pairs of sources that the reflectance is modified to be:
Here, the definition of zb, z 0, Φ, and jz) the same as defined above except that μeff and D are replaced by v̄ and μa/[v̄)-2, respectively. v̄; is the asymptotic exponential decay constant predicted by the higher order P3 approximation to the equation of radiative transport and defined in Hull and Foster .
Same as previously published studies, the tissue optical properties, i.e. the reduced scattering (μs′) and absorption (μa) coefficients, were assumed to have the form: μs′ = A(λ/λ 0)-B and μa = Σciεi(λ), respectively. Here λ is the light wavelength in nm, λ 0 is chosen to be 1 nm so that A has the same unit μs′ (i.e. cm-1), and ci and εi are the concentration and extinction coefficient of the ith chromophore, respectively. The primary chromophores in our analysis were oxyhemoglobin (HbO2), deoxyhemoglobin (Hb), lipid, and water although water and lipid absorptions are negligible at the range of separations used in this study and their absorption measured results did not affect the measurements of THC and StO2 as our previous study . The extinction coefficients of HbO2, Hb, lipid and water were obtained from the literature . A nonlinearly constrained optimization function, FMINCON implemented in MATLAB (The MathWorks, Inc., Natick, Massachusetts), was used to globally fit the data. The multi-wavelength algorithm extracts tissue scattering parameters (A and B) and chromophore concentrations (cHbO2, cHb, and clipid, and cwater). Tissue total hemoglobin concentration (THC) and blood oxygen saturation (StO2) were calculated from these quantities (e.g. THC = cHbO2 + cHb, StO2 = cHbO2/THC).
2.3 Tissue phantom preparation
We used tissue-simulating phantoms containing intact human erythrocytes, prepared according to the method of Hull et al. , at a volume of fraction of ~0.8% for StO2 validation and at various volume fractions of ~0.6% to ~2.4% for THC validation. Blood drawn from two healthy volunteers who had given informed consent was collected in tubes containing sodium heparin, mixed with an approximately equal volume of phosphate buffer solution (PBS, pH=7.4) to centrifuge, and rinsed two to three times in PBS. Finally centrifuged erythrocytes were added into a solution of ~1 % diluted Liposyn II (Abbott Labs, North Chicago, Illinois) in PBS. This solution was gently stirred using a heated stirplate (Model 460, Corning Inc., Corning, New York) to maintain a constant temperature of 37°C and homogeneity. The StO2 and pO2 of the phantom solution were measured simultaneously over the course of its deoxygenation using DRS and a microcathode pO2 electrode (Strathkelvin Instruments Limited, Scotland), respectively. Deoxygenation was accomplished by adding ~1 ml of dry yeast. The pO2 electrode was precalibrated in air-saturated water for 100% air (pO2 ~ 159 mmHg) by bubbling air into a saline solution for 15 to 20 minutes and in an oxygen-free solution for 0% oxygen (pO2 = 0 mmHg) by adding sodium sulphite to distilled water as per the manufacture’s instructions.
3.1 Validation of total hemoglobin concentration and oxygenation with tissue phantoms
Figure 2 plots the extracted StO2(%) and THC from hemoglobin phantoms using P1 and P3 approximations. The measured blood oxygen saturation (StO2(%)) is plotted versus pO2 in Fig. 2(a). Studies were performed in duplicate using blood drawn from two different subjects, respectively. The oxygen dissociation curve or Hill curve of human blood  is shown (dash line) for comparison. Compared to this Hill curve, the extracted StO2 values fit well (the difference is < 5%) for pO2 greater than ~15 mmHg (StO2=29%) but overestimated for pO2 less than 15 mmHg. It shows good agreement between P1 and P3 approximation for subject 1. In subject 2, extracted StO2(%) from P1 approximation is slightly less than that from P3 approximation. The difference is ~5% but increased up to 10 to 15% for StO2 equal to ~20–30%, then increased up to 35 to 40% for StO2 less than 10%.
The stability and accuracy of total hemoglobin concentration (THC) measurements using both P1 and P3 approximations are shown in Fig. 2(b). Each data point plot the mean value and standard deviation from 10 measurements. The stability of StO2 and μs′ was also validated in these measurements of five different THC values that the coefficient of variation (i.e. standard deviation divided by the mean) is less than 0.5% from P1 approximation and is 0.5 to 1% from P3 approximation. The accuracy (i.e. measured valued divided by the true value) of StO2 under this experimental situation was calculated to be 97 to 100% and 93 to 100% from P1 and P3 approximations, respectively. The accuracy of μs′ was calculated to be 101 to 108% from P1 approximation and 121 to 123% from P3 approximation. Overall, P1 and P3 approximations gave similar results in StO2 and THC measurements. Higher μs′ values were obtained from P3 approximation.
3.2 In vivo diffuse reflectance spectroscopy (DRS) measured higher total hemoglobin concentration and lower oxygen saturation of premalignant and malignant colonic tissues than normal colonic tissues
In vivo diffuse reflectance spectroscopy (DRS) data from a total of 27 patients (2 normal, 17 polyp, and 8 cancer patients) were analyzed using both P1 and P3 approximations to determine the total hemoglobin concentration (THC) and oxygen saturation (StO2). Similar to results in tissue phantoms, comparable StO2 and THC values were obtained using either P1 or P3 approximation. However, only results using P1 approximation are reported here for patient data because in some patients we are unable to extract the values using P3 approximation due to higher fitting error. During data fitting, we have applied the wavelength selection criteria to minimize the fitting error as described in our previous study . All polyp patients except one have pair measurements on both polyp and the corresponding adjacent normal sites. Pair measurements on both cancerous and the corresponding adjacent normal sites were taken in all cancer patients. Immediately after DRS, polyps or tumors were removed for pathology examination. Among a total of 19 resected polyps, 17 is pathologically diagnosed as tubular adenoma, 1 is inflammatory tubular adenoma, and 1 is tubular adenoma with moderate dysplasia. All resected cancerous tissues were pathologically diagnosed as adenocarcinoma. The average results and the ranges of DRS measured THC and StO2 from 2 normal patients and 24 disease (16 polyp and 8 cancer) patients that have pair measurements on both normal and disease sites are presented in Table 1. For each patient, one to two measurements per site and one to three sites per tissue type (i.e. normal, polyp, or cancer) were taken. Thus, the standard error of the mean (SEM) per tissue type (i.e. disease or normal) and per patient was calculated from one to five measurements and listed in Table 1. There are only two normal patients in this study so that the two values listed in the range of measured THC and StO2 are the results from these two subjects.
The mean±SEM values of THC and StO2 from normal sites of normal, polyp, and cancer patients are similar that no statistically difference between these values is detected (p values > 0.1). The different values between the mean THC and StO2 values of adjacent normal sites of cancer patients and those of normal and polyp patients are possible due to small sample size (n=8) and inter- and intra-heterogeneity among patients that can be seen in Fig. 3 shown later. The THC and StO2 results of normal sites from 8 cancer and 16 polyp patients have a wide range of values. The mean±SEM of THC and StO2 from all normal sites (n=26) is 93.4±17.1 μM and 67.2±3.7%, respectively.
Compared to measurements on normal sites of polyp and cancer patients, THC and StO2 on disease sites tended to increase and decrease, respectively. The decreased StO2 for malignant tumors, the decreased StO2 for polyps, and the increased THC for malignant tumors are statistically significant (p values are 0.004, 0.03 and 0.05, respectively). Figure 3 compared the results of disease (square) and adjacent normal sites (circle) from cancer (Figs. 3(a) and 3(c)) and polyp patients (Figs. 3(b) and 3(d)). The error bars plotted in Fig. 3 are the standard deviation of one to five measurements per patient as described earlier. In 8 cancer patients, 7 patients have higher THC and lower StO2 on the malignant sites than those on the adjacent normal sites. There are 7 and 11 out of 16 polyp patients having higher THC and lower StO2 on the premalignant sites than those on the adjacent normal tissues. Furthermore, significant inter- and intra-heterogeneity were observed that THC and StO2 measurements have a wide range of values on both normal and disease sites.
3.3 Disease to normal ratio of THC and StO2 measurements
To further demonstrate our findings that THC increased and StO2 decreased during tumorogenesis from normal to malignant tissues, we plotted the ratio of disease to normal THC, thus the relative total hemoglobin concentration (rTHC), and disease to normal StO2, thus the relative oxygen saturation (rStO2), in Fig. 4. The rTHC and rStO2 were shown in log scale. Except cancer patient 5 (patient c5) and approximately half of polyp patients, all other patients have shown higher THC values on the disease site than the normal site that rTHC is greater than 1 (Figs. 4(a) and 4(b)). Similarly except cancer patient 4 (patient c4) and polyp patients 1, 3, 4 and 11 (patients p1, p3, p4, and p11), all other patients have shown lower StO2 values on disease sites than normal sites that rStO2 is less than 1 (Figs. 4(c) and (d)). The mean±SEM values of rTHC and rStO2 from polyp and cancer patients are listed in Table 2 that polyp-to-normal tissue has mean rTHC and rStO2 equal to 3.2±1.1 and 0.7±0.1, respectively, and malignant-to-normal tissue has mean rTHC and rStO2 equal to 4.4±1.9 and 0.5±0.1, respectively. Although the mean rTHC and rStO2 values of malignant tissues tend to be higher and lower than those of polyps, respectively, they are not statistically different (p values = 0.5 and 0.1, respectively).
3.4 No correlation between THC and StO2 measurements and polyp sizes
Figure 5 plots the relation of THC (Fig. 5(a)) and StO2 (Fig. 5(b)) results versus the polyp size on 15 polyp measurements from 13 polyp patients. The polyp size was assessed based on endoscopic estimate and pathological report and is range from 0.2 to 2 cm. No correlation was found between DRS measured THC and StO2 and polyp size that p values are 0.48 and 0.18, respectively. The polyps that were diagnosed as inflammatory tubular adenoma and tubular adenoma with moderate dysplasia do not show bigger polyp sizes (0.5 and 1 cm, respectively) than others that were diagnosed as tubular adenoma. Their corresponding THC and StO2 values are not significantly higher and lower, respectively, than others neither although the polyp with moderate dysplasia has relatively high THC value (333.67 μM).
Malignant tumors are characterized by increased microvasculature, hence increased blood content, and poor oxygenation. Studies have shown that hypoxia stress induced angiogenesis . Tumor angiogenic marker such as VEGF is shown to be an important independent prognostic indicator in CRC patients [5, 6]. However, reports on noninvasive measurements of both blood content and oxygen level of human colon have been limited. In this study, we have demonstrated rapid, noninvasive, simultaneous measurements of total hemoglobin concentration and oxygen saturation in normal, premalignant, and malignant colonic tissues in vivo. The results exhibit substantial inter- and intra-patient heterogeneity within normal and disease (premalignant and malignant) sites of colon walls. We have also observed an increased THC and decreased StO2 during colon carcinogenesis from normal to malignant tumors as expected.
Our results of mean±SEM StO2 of 72.0±0.8%, 70.7±4.6%, and 60.7±7.7% in normal colonic tissue sites of normal, polyp, and cancer patients, respectively, of 51.3±7.0% in premalignant adenomatous polyps, and of 26.4±6.1% in malignant tumors agree with other published results. For normal colon measurements, Friedland et al. used a commercial oximetry probe based on endoscopic visible light spectroscopy to measure chronic mesenteric ischemia . The results from two studies have shown the mean±standard deviation oxygen saturation of normal colorectal mucosa from 40 and 25 patients is 72±3.5% and 73±5%, respectively, and ranges from 61 to 86% [31, 32]. Zonios et al. used similar spectroscopy method with a fiber optic probe and diffuse model based analysis to perform measurements from 13 patients that normal colon has mean±standard deviation StO2 of 59±8% with a range of 40 to 90% . For premalignant colonic tissues, Zonios et al. measured StO2 of 63±10% with a range of 40 to 90% . For malignant tumor measurements, Gatenby et al. measured oxygen tension in vivo in multiple lesions and normal adjacent tissues using computed tomography guided electrode probe . Oxygen tensions (pO2) at 4 to 30 mmHg (≈ StO2 =2 to 68%) were reported in human malignant colon adenocarcinoma that are ~7 times lower than those reported in the adjacent normal sites. Compared to results by Zonios et al. that normal colon and adenomatous polyp have similar mucosal StO2 values, we have detected slightly higher and lower mean StO2 in all normal sites (67.2±3.7% versus 59±8%) and adenomatous polyps (50.3±6.9% versus 63±10%), respectively. This difference is in part due to reflectance signals collected from slightly different detected tissue area and depth. Zonios et al. used a source-detector separation ~0.55 mm and wavelength range of ~400 to 700 nm that collected optical signals mainly from the shallower layer of the colon wall. We used multiple source-detection separations ranging from 0.6 mm to ~2.5 mm and a wavelength range of ~600 to 800 nm that collected signals from a deeper tissue layer, thus a wider tissue area for averaging signals.
Reported THC of normal and diseased colonic tissues has a wide range. Zonios et al. reported ~5 to 35 mg/dl (≈ 3 to 21 μM) in normal colons and ~5 to 35 mg/dl (≈ 3 to 21 μM) in adenomatous polyps . Roy et al. measured the THC from 222 patients (175 normal, 47 adenomas) undergoing colonoscopy using a polarization gated fiber optic probe for better selection of penetration depths at 100, 130 and 170 μm below tissue surface . They measured that the midtransverse colon, for example, has mean THC value of 7.65 g/l (~ 464 μM) and it increased up to 75.3% at adenoma sites. Our measured results of mean THC of ~110 and ~137 μM in normal and premalignant colonic tissues, respectively, lie within reported ranges. The ratio of adenoma to normal mean THC in our study (rTHC~137/110=1.24) is slightly lower that that reported in Roy et al. (i.e. 1.7±0.2 to 1.8±0.3). However, we did observe higher ratios with a broader range (i.e. mean rTHC is 3.2±1.1) when we calculated the ratio based on individual patient.
That we did not find correlation between polyp size and THC and StO2 is possibly due to small sample size and because the premalignant tissues measured in this study belong to similar pathological category: 17 of 19 resected polyps were tubular adenoma, 1 inflammatory tubular adenoma, and 1 tubular adenoma with moderate dysplasia. Roy et al. categorized polyps having size ≥ 10 mm or presence of high grade dysplasia as advanced adenomas and categorized adenoma as nonadvanced adenoma. Although half of our measured polyps have size of 10 mm or greater, none of them has presence of high grade dysplasia. Further study with a bigger sample size is necessary to test if THC and StO2 are correlated with the grade of the disease and/or with the polyp size.
We thank Dr. Jarod Finlay who provided MATLAB codes of the hybrid P3 approximation and Professor Yau-Hui Wei who provided the oxygen electrode probe. We acknowledge the technical assistance by Mr. Han-Wen Guo in preparing blood samples. This work is supported by a new faculty start-up fund from National Yang-Ming University, the “Aim for Top University Plan” from the Ministry of Education of Taiwan, and grants NSC 95-2112-M-010-002 from the National Science Council of Taiwan and CI-96-8 from Yen Tjing Ling Medical Foundation.
References and links
4. S. Valable, B. Lemasson, R. Farion, M. Beaumont, C. Segebarth, C. Remy, and E. L. Barbier, “Assessment of blood volume, vessel size, and the expression of angiogenic factors in two rat glioma models: a longitudinal in vivo and ex vivo study,” NMR Biomed. 21, 1043–1056 (2008). [CrossRef] [PubMed]
5. F. De Vita, M. Orditura, E. Lieto, S. Infusino, F. Morgillo, E. Martinelli, P. Castellano, C. Romano, F. Ciardiello, G. Catalano, C. Pignatelli, and G. Galizia, “Elevated perioperative serum vascular endothelial growth factor levels in patients with colon carcinoma,” Cancer 100, 270–278 (2004). [CrossRef] [PubMed]
6. G. Lurje, W. Zhang, A. M. Schultheis, D. Yang, S. Groshen, A. E. Hendifar, H. Husain, M. A. Gordon, F. Nagashima, H. M. Chang, and H. J. Lenz, “Polymorphisms in VEGF and IL-8 predict tumor recurrence in stage III colon cancer,” Ann. Oncol. 19, 1734–1741 (2008). [CrossRef] [PubMed]
7. H. Hurwitz, L. Fehrenbacher, W. Novotny, T. Cartwright, J. Hainsworth, W. Heim, J. Berlin, A. Baron, S. Griffing, E. Holmgren, N. Ferrara, G. Fyfe, B. Rogers, R. Ross, and F. Kabbinavar, “Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer,” N. Engl. J. Med. 350, 2335–2342 (2004). [CrossRef] [PubMed]
9. P. Vaupel, F. Kallinowski, and P. Okunieff, “Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors: a review,” Cancer Res. 49, 6449–6465 (1989). [PubMed]
10. A. Mayer, A. Wree, M. Hockel, C. Leo, H. Pilch, and P. Vaupel, “Lack of correlation between expression of HIF-1alpha protein and oxygenation status in identical tissue areas of squamous cell carcinomas of the uterine cervix,” Cancer Res. 64, 5876–5881 (2004). [CrossRef] [PubMed]
11. S. M. Evans, K. D. Judy, I. Dunphy, W. T. Jenkins, W. T. Hwang, P. T. Nelson, R. A. Lustig, K. Jenkins, D. P. Magarelli, S. M. Hahn, R. A. Collins, M. S. Grady, and C. J. Koch, “Hypoxia is important in the biology and aggression of human glial brain tumors,” Clin. Cancer Res. 10, 8177–8184 (2004). [CrossRef] [PubMed]
12. T. M. Busch, S. M. Hahn, E. P. Wileyto, C. J. Koch, D. L. Fraker, P. Zhang, M. Putt, K. Gleason, D. B. Shin, M. J. Emanuele, K. Jenkins, E. Glatstein, and S. M. Evans, “Hypoxia and Photofrin uptake in the intraperitoneal carcinomatosis and sarcomatosis of photodynamic therapy patients,” Clin. Cancer Res. 10, 4630–4638 (2004). [CrossRef] [PubMed]
14. G. J. Hutchison, H. R. Valentine, J. A. Loncaster, S. E. Davidson, R. D. Hunter, S. A. Roberts, A. L. Harris, I. J. Stratford, P. M. Price, and C. M. West, “Hypoxia-inducible factor 1alpha expression as an intrinsic marker of hypoxia: correlation with tumor oxygen, pimonidazole measurements, and outcome in locally advanced carcinoma of the cervix,” Clin. Cancer Res. 10, 8405–8412 (2004). [CrossRef] [PubMed]
15. D. Furlan, N. Sahnane, I. Carnevali, R. Cerutti, F. Bertoni, I. Kwee, S. Uccella, V. Bertolini, A. M. Chiaravalli, and C. Capella, “Up-regulation of the hypoxia-inducible factor-1 transcriptional pathway in colorectal carcinomas,” Hum. Pathol. 39, 1438–1494 (2008). [CrossRef]
17. B. J. Tromberg, B. W. Pogue, K. D. Paulsen, A. G. Yodh, D. A. Boas, and A. E. Cerussi, “Assessing the future of diffuse optical imaging technologies for breast cancer management,” Med. Phys. 35, 2443–2451 (2008). [CrossRef] [PubMed]
18. T. Durduran, G. Yu, M. G. Burnett, J. A. Detre, J. H. Greenberg, J. Wang, C. Zhou, and A. G. Yodh, “Diffuse optical measurement of blood flow, blood oxygenation, and metabolism in a human brain during sensorimotor cortex activation,” Opt. Lett. 29, 1766–1768 (2004). [CrossRef] [PubMed]
19. H. W. Wang, T. C. Zhu, M. E. Putt, M. Solonenko, J. Metz, A. Dimofte, J. Miles, D. L. Fraker, E. Glatstein, S. M. Hahn, and A. G. Yodh, “Broadband reflectance measurements of light penetration, blood oxygenation, hemoglobin concentration, and drug concentration in human intraperitoneal tissues before and after photodynamic therapy,” J. Biomed. Opt. 10, 14004 (2005). [CrossRef] [PubMed]
20. G. Yu, T. Durduran, G. Lech, C. Zhou, B. Chance, E. R. Mohler 3rd, and A. G. Yodh, “Time-dependent blood flow and oxygenation in human skeletal muscles measured with noninvasive near-infrared diffuse optical spectroscopies,” J. Biomed. Opt. . 10, 024027 (2005). [CrossRef] [PubMed]
21. U. Sunar, H. Quon, T. Durduran, J. Zhang, J. Du, C. Zhou, G. Yu, R. Choe, A. Kilger, R. Lustig, L. Loevner, S. Nioka, B. Chance, and A. G. Yodh, “Noninvasive diffuse optical measurement of blood flow and blood oxygenation for monitoring radiation therapy in patients with head and neck tumors: a pilot study,” J. Biomed. Opt. 11, 064021 (2006). [CrossRef]
22. G. Zonios, L. T. Perelman, V. Backman, R. Manoharan, M. Fitzmaurice, J. Van Dam, and M. S. Feld, “Diffuse reflectance spectroscopy of human adenomatous colon polyps in vivo,” Appl. Opt. 38, 6628–6637 (1999). [CrossRef]
23. H. W. Wang, J. C. Finlay, K. Lee, T. C. Zhu, M. E. Putt, E. Glatstein, C. J. Koch, S. M. Evans, S. M. Hahn, T. M. Busch, and A.G Yodh, “Quantitative comparison of tissue oxygen and motexafin lutetium uptake by ex vivo and noninvasive in vivo techniques in patients with intraperitoneal carcinomatosis,” J. Biomed. Opt. 12, 034023 (2007). [CrossRef] [PubMed]
24. E. L. Hull and T. H. Foster, “Steady-state reflectance spectroscopy in the P3 approximation,” J. Opt. Soc. Am. A. 18, 584–599 (2001). [CrossRef]
25. J. K. Fisher, “Normal colon wall thickness on CT,” Radiology 145, 415–418 (1982). [PubMed]
28. J. C. Finlay and T. H. Foster, “Hemoglobin oxygen saturations in phantoms and in vivo from measurements of steady-state diffuse reflectance at a single, short source-detector separation,” Med. Phys. 31, 1949–1959 (2004). [CrossRef] [PubMed]
29. S. Prahl, “Optical Properties Sepctra,” http://omic.ogi.edu/spectra/index.html.
30. E. L. Hull, M. G. Nichols, and T. H. Foster, “Quantitative broadband near-infrared spectroscopy of tissue-simulating phantoms containing erythrocytes,” Phys. Med. Biol. 43, 3381–3404 (1998). [CrossRef] [PubMed]
31. S. Friedland, D. Benaron, S. Coogan, D. Y. Sze, and R. Soetikno, “Diagnosis of chronic mesenteric ischemia by visible light spectroscopy during endoscopy,” Gastrointest. Endosc. 65,294–300 (2007). [CrossRef]
32. S. Friedland, D. Benaron, I. Parachikov, and R. Soetikno, “Measurement of mucosal capillary hemoglobin oxygen saturation in the colon by reflectance spectrophotometry,” Gastrointest. Endosc. 57, 492–497 (2003). [CrossRef] [PubMed]
33. R. A. Gatenby, L. R. Coia, M. P. Richter, H. Katz, P. J. Moldofsky, P. Engstrom, D. Q. Brown, R. Brookland, and G. J. Broder, “Oxygen tension in human tumors: in vivo mapping using CT-guided probes,” Radiology 156, 211–214 (1985). [PubMed]
34. H. K. Roy, A. Gomes, V. Turzhitsky, M. J. Goldberg, J. Rogers, S. Ruderman, K. L. Young, A. Kromine, R. E. Brand, M. Jameel, P. Vakil, N. Hasabou, and V. Backman, “Spectroscopic microvascular blood detection from the endoscopically normal colonic mucosa: biomarker for neoplasia risk,” Gastroenterology 135, 1069–1078 (2008). [CrossRef] [PubMed]