Journal of Cancer Research and Therapeutics

: 2008  |  Volume : 4  |  Issue : 1  |  Page : 26--36

An overview on applications of optical spectroscopy in cervical cancers

Murali Krishna Chilakapati1, GD Sockalingum2, MS Vidyasagar3, M Manfait2, Donald J Fernanades3, BM Vadhiraja3, K Maheedhar4,  
1 Division of Laser Spectroscopy, Manipal Life Science Center, Manipal University, Manipal, Karnataka, India
2 Unité MéDIAN, CNRS UMR 6142, UFR Pharmacie, IFR 53, Université de Reims Champgane-Ardenne, 51 rue Cognacq - Jay, Reims Cedex, France
3 Department of Radiotherapy and Oncology, Shirdi Sai Baba Cancer Hospital, Kasturba Medical College, Manipal University, Manipal, Karnataka, India
4 Division of Laser Spectroscopy, Manipal Life Science Center and Department of Radiotherapy and Oncology, Shirdi Sai Baba Cancer Hospital, Kasturba Medical College, Manipal University, Manipal, Karnataka, India

Correspondence Address:
Murali Krishna Chilakapati
Chilakapati Laboratory, ACTREC, TMC, Kharghar, Sector «SQ»22«SQ», Navi Mumbai - 410 208, Maharastra


Despite advances in the treatment modalities, cervical cancers are one of the leading causes of cancer death among women. Pap smear and colposcopy are the existing screening methods and histopathology is the gold standard for diagnosis. However, these methods have been shown to be prone to reporting errors, which could be due to their subjective interpretation. Radiotherapy is the mainstay of treatment for the locally advanced stages of cervical cancers. The typical treatment regimen spans over 4 months, from the first fraction of radiation to clinical assessment of tumor response to radiotherapy. It is often noticed that due to intrinsic properties of tumors, patients with the same clinical stage and histological type respond differently to radiotherapy. Hence, there exists a need for the development of new methods for early diagnosis as well as for early prediction of tumor radioresponse. Optical spectroscopic methods have been shown to be potential alternatives for use in cancer diagnosis. In this review, we provide a brief background on the anatomy and histology of the uterine cervix and the etiology of cervical cancers; we briefly discuss the optical spectroscopic approach to cervical cancer diagnosis. A very brief discussion on radiation therapy and radiation resistance is also provided. We also share our experiences with the Raman spectroscopic methodologies in cervical cancer diagnosis as well as in the prediction of tumor radioresponse.

How to cite this article:
Chilakapati MK, Sockalingum G D, Vidyasagar M S, Manfait M, Fernanades DJ, Vadhiraja B M, Maheedhar K. An overview on applications of optical spectroscopy in cervical cancers.J Can Res Ther 2008;4:26-36

How to cite this URL:
Chilakapati MK, Sockalingum G D, Vidyasagar M S, Manfait M, Fernanades DJ, Vadhiraja B M, Maheedhar K. An overview on applications of optical spectroscopy in cervical cancers. J Can Res Ther [serial online] 2008 [cited 2019 Oct 14 ];4:26-36
Available from:

Full Text

The cervix is an extension of the body of the uterus. The upper end communicates with the uterine body via the internal os and the lower end opens into the vagina through the external os. Except for the lower part, the uterine cervix is lined by a single-layered, columnar epithelium with tubular glands. The lower region is covered by nonkeratinizing stratified squamous epithelium. The squamocolumnar junction - between the columnar secretory epithelium of the endocervical canal and the stratified squamous covering of the ectocervix - has the greatest clinical significance as it is most common site for epithelial abnormalities that may progress to malignancy.

Cervix cancers are the second most common cancers among women (15% of female cancers in developing countries) and the seventh in frequency overall. Estimates suggest that approximately 4,93,000 patients were diagnosed with cervical cancer and 2,74,000 died from the disease in the year 2002. [1] In developing countries, cervical cancer accounts 15% female cancers out of 83% new cases. Among this women below 65 years age group carry a risk of 1.5%. In developed countries, cervical cancer accounts for only 3.6% of all new cancers, with a cumulative risk of 0.8% for the same age-group. The major risk factor for cervical cancers is human papilloma virus (HPV) infection, especially with strains like HPV 16, 18, 31, 33, and 45, which are termed as high risk factor strains. Other known risk factors include: smoking, HIV infection, low socioeconomic status, chlamydia infection, oral contraceptives, and multiple pregnancies. [2]

Pap smear and colposcopy are widely-used screening methods for the detection of cervical cancers. Eighty percent of all cervical cancer deaths are reported in developing countries, where these screening tests are not routinely practiced. [3] This emphasizes the importance of effective screening methods and early clinical detection techniques in cervical cancer management. However, the existing screening techniques have also been shown to suffer from high false negative/positive results, which could be attributed to the subjective interpretations of these tests. [4],[5] Histopathology, the gold standard of diagnosis, is a time-consuming process and has also been shown to be influenced by subjective interpretation.

Radiotherapy is the treatment of choice for the locally advanced stages of cervical cancers. The radiotherapy regimen, from the first fraction of treatment till clinical evaluation of tumor radioresponse, spans over 4 months. A typical radiotherapy regimen is as follows: a patient with stage IIB to stage IV disease is subjected to external-beam radiotherapy (EBRT) of 45 Gy in 20 fractions over a period of 4 weeks, using a linear accelerator. The patient is then allowed to rest for 2 weeks for parametrial regression, after which two doses of remote afterloaded high dose rate intracavitary brachytherapy (HDR) of 8.5 Gy to point A is given once a week. After 4 weeks of rest the patient is subjected to clinical assessment by per vaginal and per rectal examination 'to evaluate' tumor radioresponse; that is, conventionally, tumor response to radiotherapy is evaluated after 4 months from patient is exposed to first fraction of radiation treatment. Radiation resistance is a well-known and a serious hurdle in radiotherapy of cancers. Intrinsic factors such as DNA aneuploidy, S-phase fraction and proliferation kinetics, tumor vascularity and hypoxia, and glutathione content are believed to influence the radiation resistance of a tumor. [6],[7],[8] Under the influence of these intrinsic factors, tumors of the same clinical stage and histological type often exhibit differential radioresponse. The tumor response to radiation therapy is graded based on the degree of shrinkage of the volume of the tumor, which is, as mentioned above, assessed at the end of the treatment. A 100% shrinkage in the volume indicates complete response, shrinkage in volume of 65% and above is considered as partial response, and shrinkage in volume of less than 65% is assumed to indicate no response. [9],[10] Very few studies have been reported in the literature on prediction of tumor response using radiobiological, magnetic resonance, and NMR spectroscopy methods. [6],[11],[12],[13],[14] Radiobiological studies indicate that prediction of tumor response to radiation could be assessed by estimation of glutathione (GSH) levels. [6] Though these studies do demonstrate the feasibility of prediction of tumor radioresponse, so far there are no prescribed methods for use in clinical practice.

From the above discussion it is clear that conventional screening methods have high rates of reporting errors, and there is no established method to predict tumor radioresponse at an early stage of the treatment. Hence, there exists a need for the development of new methods: 1) for detecting malignancy at an early stage and 2) for predicting the tumor radioresponse at an early stage of the treatment so that treatment protocols can be individualized.

Optical spectroscopic techniques, fluorescence, [15],[16],[17],[18],[19],[20],[21],[22],[23],[24] FTIR [25],[26],[27],[28],[29],[30] and Raman, [31],[32],[33],[34],[35],[36] which are sensitive to changes at the level of biomolecular composition of samples, have been gaining considerable importance in biomedical applications, including diagnosis of cancer. In case of autofluorescence, the sample is impinged by suitable radiation, causing excitation of intrinsic fluorophores. These molecules emit photons on their return to the ground state. FTIR and Raman are the two complementory vibrational spectroscopic techniques. Infrared (IR) spectroscopy is an absorption process, whereas Raman spectroscopy is an inelastic scattering phenomenon. Fluorescence and FTIR are more sensitive and are popular methods in optical diagnosis. In the case of Raman scattering, molecules are excited to a virtual state on interaction with photons. The vast majority of these molecules deexcite back to the initial vibrational energy level, which results in release of photons of the same energy and wavelength - Rayleigh scattering. On the other hand a very tiny fraction of molecules deexcite to a different vibrational state. This phenomenon is known as 'Raman scattering.' A shift of scattering towards longer wavelengths is called the 'Stokes effect' and a shift towards shorter wavelengths leads to 'anti-Stokes.' Nevertheless Stokes lines are the more predominant and are commonly observed in Raman spectroscopy. The energy level transitions of various photophysical processes are depicted in [Figure 1]. FTIR and Raman spectroscopy provides rich information through 'molecular fingerprint.' Though, as described earlier, Raman scattering is an inherently weak process, Raman spectroscopy has certain distinct advantages over the other two popular methods, which include: 1) virtually no sample preparation ( vs FTIR), 2) no strict dependence on wavelengths ( vs fluorescence), 3) use of less harmful NIR radiations ( vs fluorescence), 4) adaptability to in vivo / in situ measurements ( vs FTIR), and 5) easy extraction of information from sharp spectral features ( vs fluorescence and FTIR). However, as been stated earlier, a major drawback of Raman scattering is that it is an inherently weak process. To overcome this, several improvisations have been developed over conventional Raman spectroscopy; these include: resonance Raman, surface-enhanced Raman scattering (SERS), and Raman microspectroscopy.

 Optical Spectroscopy - Cervical Cancer

Fluorescence spectroscopy

To the best of our knowledge, Alfano et al . [37] were the first to measure autofluorescence spectra of normal and malignant tissues. The group led by R.R. Kortum has been carrying out extensive investigations on fluorescence spectroscopy in normal and pathological conditions of the uterine cervix. [38],[39],[40],[41],[42],[43],[44],[45],[47],[48],[49],[50],[51],[52],[53],[54] They have demonstrated the classification of normal and pathological cervical tissues at different excitation wavelengths as early as 1993. [38] The same group has reported that an increase in NADH fluorescence and a decrease in collagen fluorescence provided clinically significant differences between normal and dysplastic tissues [45] ; this corroborates the findings of Feld's group, which proposed that NAD(P)H and collagen could be used as quantitative fluorescence biomarkers for in vivo detection of dysplasia. [46] In a combined fluorescence and reflectance spectroscopy study they have shown that fluorescence emission spectra in 330-360 nm and 460-470 nm provided the best discrimination among normal, precancerous, and cancerous tissues compared to reflectance spectroscopy as well as a combination of both. [52] More recently they have developed a multispectral digital colposcope (MDC). [54] MDC is an unique multispectral imaging system that is a modification of the standard colposcope with a video camera adapter to measure reflectance and fluorescence images of the uterine cervix using an color video camera. Feld and his coworkers, the other major group in the field of laser-induced fluorescence (LIF), have demonstrated the potential of trimodal spectroscopy (intrinsic fluorescence, diffuse reflectance, and light scattering) for in vivo detection of cervical precancerous changes. [55] Recently, in another study, the accuracy of three integrated optical measurements - LIF, white light diffuse reflectance spectroscopy, and video imaging - in the detection of cervical disorders was investigated by Schomacker et al . [56] Several groups have investigated the potential of fluorescence spectroscopy in cervical cancer diagnosis. Gupta and his coworkers have demonstrated the efficacy of nitrogen LIF in cervical cancer diagnosis. [57] The group led by Kartha has demonstrated the potential of fluorescence spectroscopy at 325 nm excitation in discriminating between normal and malignant cervical tissues. [58] The novelty of their study was the comparison of gross tissue spectral features with the fluorescence emitted by individual fluorophores of the tissue homogenates, which was measured using high-performance liquid chromatography-LIF (HPLC-LIF) instrumentation. Recently, a correlative study between fluorescence spectra and detailed histopathological protocol of various parameters (such as sample thickness, type of epithelium, orientation, and location of the biopsy, etc.) demonstrated that fluorescence spectroscopy could classify disorders of the cervix with a high sensitivity and specificity. [59],[60] Recent time-resolved confocal fluorescence experiments on monolayered cell cultures, using 365 nm excitation, demonstrated that time-resolved autofluorescence decays of free and bound NADH signals are the sensitive indicators of cellular metabolism. [61] Alvarez et al . reported that a combination of optical spectroscopy with colposcopy provided a clinically meaningful increase in the detection of CIN I, III in women referred for the evaluation of mildly abnormal cytology results. [62] Martin et al . characterized the spectroscopic changes that take place during neoplastic progression of cervical cancer using organotypic epithelial raft culture as an in vitro model of cervical tissue. [63] Werner et al . demonstrated that spectroscopy combined with cervical cytology is equally sensitive and twice as specific as HPV testing alone in identifying high-grade cervical neoplasia. [64]

FTIR spectroscopy

To the best of our knowledge Wong et al . were the first to record the IR spectra of normal connective tissue and the normal and malignant epithelial tissue of the cervix. They proposed that this technique could be used for mass screening. [65] A few other groups had demonstrated that IR spectroscopy could be used not only for the diagnosis of cervical cancers but also for characterizing molecular abnormalities during progression of the disease (CIN I-III). [66],[67] In a comparative study, Fung et al . demonstrated that FTIR has a better false-negative rate and negative predictive value than standard Pap smears. [68] Cohenford et al . demonstrated the efficacy of the chemometric approach, using principal components analysis (PCA) for the discrimination of several conditions of the cervix. [69],[70] The group led by Diem carried out extensive FTIR spectroscopy investigations on formalin-fixed paraffin-embedded cervical tissues, cell lines, and exfoliated cells. [71],[72],[73],[74],[75],[76],[77] They found that dysplastic samples have spectral features that are about halfway between that of normal and cancerous samples. [70],[71] They demonstrated the implications of IR microspectroscopy in monitoring cell proliferation, drug response, etc., using He La cell lines [72],[73],[74] and also reported that IR spectroscopy is a highly sensitive technique for the detection and differentiation of cell types and for diagnosis. [75] They have demonstrated that a combination of FTIR microspectroscopy and multivariate spectral processing methods provide important insights into the fundamental spectral signatures of individual cells. [76] More recently, they have developed a database of IR spectral patterns for different tissue types as well as for different stages of cancer. [77] As described above, several studies have explored the feasibility of diagnosis of cervical cancers using exfoliated cell smears. As is well known, FTIR spectroscopy demands several steps of sample preparation and this can, in the end, induce artifacts if proper care is not taken. [78] The heterogeneity of real-life cell smears compounds the problem. McNaughton and his coworkers have investigated this aspect. [79],[80] Though several factors such as bacterial contamination and the presence of leukocytes and thrombocytes can influence the spectral features, they opine that it is still possible to arrive at a diagnosis by means of careful data analysis. An independent study by Shaw et al . demonstrates the same. [81] Chang et al . found that the ratio of the areas of two spectral regions between 1130-1180 cm -1 and 1180-1260 cm -1 was an exceptionally useful factor in discriminating precancerous tissues from normal tissues of the uterine cervix. [82] Mark et al . reported that the grading of neoplasia based on FTIR microspectroscopy (FTIR-MSP) and probabilistic neural network (PNN) differentiates normal from premalignant tissue with a high level of accuracy. [83],[84] Mordechai et al . demonstrated that in FTIR-MSP of formalin-fixed cervical tissues, glycogen levels could be a potential biomarker in the detection of cervical neoplasia. [85] However, this aspect is a matter of debate. In a very recent work, Walsh et al . reported that attenuated total reflection-Fourier transform IR (ATR-FTIR) coupled with PCA and subsequent linear discriminant analysis (LDA) facilitated the identification of phosphate and glycogen in the IR spectral region of 950-1200 cm -1 as potential biomarkers of abnormality. [86]

Raman spectroscopy

As mentioned earlier (introduction) section, FTIR and LIF are far more popular methods than Raman. This could be attributed to the higher sensitivity and simpler instrumentation of LIF and FTIR. Though recent advances in instrumentation (lasers and CCDs) make the recording of the Raman signal from biological samples relatively easy, the number of Raman-based studies are very few. Mahadevan et al . were the first to record the Raman spectrum of cervical tissues. [87] As early as 1998, they reported the potential of NIR Raman spectroscopy to distinguish cervical precancers from other conditions. They developed a compact fiberoptic probe to measure the in vivo Raman spectra of cervical tissue [88] and have demonstrated its potential in the detection of squamous dysplasia in normal and precancerous conditions. [89] Very recently, Lyng et al . demonstrated the potential of Raman microspectroscopy in the identification of biochemical changes in tissues during disease progression, using formalin-fixed paraffin-preserved (FFPP) tissues. [90]

 Raman Spectroscopy of Cervical Tissues: Our Approach

Diagnostic applications

Some of the major issues pertaining to biomedical application are as follows:

1. Since the cervix is easily accessible, it is possible to develop noninvasive techniques using suitable fiberoptic probes. To carry out such studies, conventional Raman spectroscopy by NIR radiation is more suitable as the NIR radiation is relatively less harmful. Since conventional spectroscopy probes a large area ( vs microspectroscopy), the spectra are more representative and, to a reasonable extent, the spectra recorded in ex vivo conditions can be extrapolated to in vivo conditions.

2. Methodologies should be relatively simple so that an untrained technician can use them. However, rigorous evaluation of models is a prerequisite before implementation in routine practice. Optical spectroscopic data are amenable to multivariate statistical tools such as artificial neural network (ANN), hierarchical cluster analysis (HCA), discriminant analysis (DA), principle components analysis (PCA), etc. Since there are several discriminating algorithms, it is necessary to explore and identify a suitable rapid, robust, and simple discriminating algorithm from the clinician's point of view.

In view of these considerations, we have developed a conventional Raman spectroscopic diagnostic model for discrimination of normal and malignant tissues of the uterine cervix using 15 normal and 24 malignant certified tissues. [91] We further evaluated the model by using 72 blinded samples, where the spectroscopists did not have prior information on the nature of the sample. A typical Raman setup assembled by us is shown in [Figure 2].

Mean Raman spectra of normal and malignant tissues is shown in [Figure 3]. The normal spectrum is characterized by broader Amide I and III and peaks at 1278, 864 and 948 cm -1 , which can be assigned to structural proteins such as collagen and elastin [Figure 3]A. The mean malignant spectrum shows sharper amide I, minor blue shift in δCH 2 band, and sharper features in the amide III region, and these bands can be assigned to biomolecules such as lipids, DNA, and noncollagenous proteins [Figure 3]B. [92],[93] These observations corroborate earlier studies. [87]

We have analyzed spectral data by PCA. Data analysis was carried out in both supervised and unsupervised modes. In the unsupervised approach, spectra from different tissue types were pooled and analyzed. This approach is necessary to verify if spectra can be classified on the lines of normal and malignant conditions. In our analysis, scores of factor 1 gave good classification among normal and malignant tissue spectra of the uterine cervix [Figure 4]A. Though unsupervised classification based on scores of factor is a widely used approach to obtain discrimination between tissue types, it is somewhat cumbersome and tedious for routine diagnosis from the clinical point of view. Therefore, we have developed a supervised approach where multiple discriminating parameters can be used to achieve an objective diagnosis. In this direction, we have built models for the normal and malignant conditions using 24 randomly picked normal and 28 malignant spectra. Exclusiveness and the representative property of these models were verified by rotating out spectra and comparing them against both the standard sets, using Mahalanobis distance and spectral residuals as discriminating parameters. [94],[95],[96] In this case, if the test spectrum and the standard set belongs to the same class, then the values of Mahalanobis distance and spectral residuals should be very low and vice versa. As an example, verification against the malignant standard set is shown in [Figure 4]B. As expected, all malignant spectra yielded lower values of Mahalanobis distance and spectral residual, whereas normal spectra gave higher values. Thus good discrimination among normal and malignant condition could be achieved, which suggests that standard sets are exclusive and representative. Further, these standard sets are evaluated by test spectra (i.e., spectra that were not a part of the standard sets). Once again, all spectra are matched against both the standard sets. As an example, evaluation against the normal standard set is shown in [Figure 4]C. Good classification among the tissue types could be obtained. This approach to data analysis is less tedious as compared to the cumbersome PCA, but it requires training. In the next step we have implemented match/mismatch 'limit test' methodology, where pretreated spectra are matched against both the standard sets with fixed inclusion/exclusion criterion for all the three discriminating parameters: scores of factor, Mahalanobis distance, and spectral residuals. In this case, spectra are classified on the basis of both match and mismatch status, i.e., a normal spectrum should match with the normal standard set and fail or mismatch with the malignant standard set. Thus, the match/mismatch status of each spectrum against both the standard sets provides objective discrimination [Table 1]. This methodology was successfully implemented to diagnose oral, [97] breast, [98] stomach, [99] and colon [100] cancers. As shown in [Table 1], malignant spectra (1-68) did not match, whereas all normal spectra (69-150) match against the normal standard set. We have obtained exactly opposite results against the malignant standard set (data not shown). Thus, we could differentiate both the tissue types objectively and unambiguously.

However, it is very essential for medical applications that any diagnostic methodology is verified and tested rigorously over a large sample size in order to establish its validity before it is routinely practiced. Hence, our discriminating methodology was further evaluated by 70 blinded samples (the nature of the sample was not revealed to the spectroscopist) using the 'limit test' methodology. To illustrate the advantages of the methodology, analysis of 20 blinded samples is shown here [Table 2]. All spectra of samples 1, 2, 4-11, 14, 15, and 17-20 match with only one standard set and do not match with another and they can be diagnosed, in a straightforward manner, as normal or malignant. In the case of sample 3, at least one spectrum matches with the malignant standard set; thus it is treated as malignant in the lines of histopathology. For samples 12, 13, and 16, most of the spectra match with the normal standard set. Thus, these tissues are treated as normal. The results obtained in the analysis show good correlation between Raman spectroscopy and histopathology.

 Predictive Assays for Response to Radiotherapy

We have also explored the feasibility of prediction of tumor radioresponse at an early stage of treatment. [101],[102] In order to achieve this we have recorded Raman spectra of cervix cancer tissues that were collected before radiation therapy (referred to as malignant) and 24 h after the patient was treated with the second fraction of external-beam radiotherapy (referred to as 2-RT). Independent unsupervised analysis of malignant and 2-RT spectra was carried out. Unsupervised PCA of malignant tissue spectra failed to provide any classification [Figure 5]A, whereas PCA of 2-RT spectra gave clear classification of responding (complete and partial response) and nonresponding conditions, based on the score of factor 1. A tendency of separation among responding conditions based on score of factor 2 could also be observed [Figure 5]B.

We have also further explored the feasibility of classification among responding conditions. In order to achieve this, we have recorded the Raman spectra of tissues that were collected 24 h after the fifth fraction of EBRT (referred to as 5-RT). [102] PCA of 5-RT tissue spectra could discriminate between responding and nonresponding conditions based on the score of factor 1, and a tendency of separation was also observed between complete response and partial response based on the score of factor 2 [Figure 5]C. The clinical assessment of tumor response of a few cases is still awaited. Nevertheless, good correlation could be seen between Raman spectroscopy and available tumor response.

 Suitability of Formalin - Fixed Tissues in Optical Diagnosis by Raman Spectroscopy

Fresh tissues in saline would be ideal for optical spectroscopy, but this often causes major constraints in sample procurement and handling. Because of the scarcity and other associated problems with fresh tissues and the availability of ex vivo samples, it would be very useful to evaluate the suitability of fixed samples for optical pathology since they are more readily available and accessible for such studies. Paraffin-embedded tissues are the most widely used procedure for sample storage in histopathology. In order to evaluate the suitability of ex vivo handled tissues from the Raman spectroscopy point of view, we carried out pilot Raman and FTIR microspectroscopic study of formalin-fixed normal, malignant, and 2-RT tissues. [103] Raman and FTIR spectra of these tissues were recorded using commercial Raman microspectrometer (LabRam, Jobin-Yvon - Horiba, France) and FT-IR imaging system (SPOTLIGHT, Perkin-Elmer, France) coupled to a FT-IR spectrometer (Spectrum 300, Perkin-Elmer, France), respectively. Raman and FTIR spectra exhibit large differences for normal and malignant tissues and subtle differences are seen between malignant and 2-RT tissues. The major differences between the mean Raman spectra of malignant and normal tissue are the relatively stronger peaks of δCH 2 and amide III in the 1000-1200 cm -1 and 600-800 cm -1 regions and weak peaks in the 800-1000 cm -1 region [Figure 6]A. As mentioned above, the differences between the malignant and 2-RT spectra are very minute [Figure 6]B. Typical mean FTIR spectra of normal and malignant epithelia also exemplify significant differences as shown in [Figure 7]A, whereas differences among malignant and 2-RT spectra are very minute [Figure 7]B. Spectral data were analyzed by two-step PCA. In the first step of the cluster analysis, spectra from all tissue types (normal, malignant, and 2-RT) were pooled and analyzed. This resulted in two clusters corresponding to normal and malignant + 2-RT [Figure 8]A, B. Unsupervised analysis of malignant and 2-RT spectra was carried out as a second step. In this analysis, Raman spectra gave better classification among malignant and 2-RT compared to FTIR spectra [Figure 9]A, B. This indicates the feasibility of discriminating 2-RT tissues from malignant despite the minute biochemical differences. However, these results could not be pursued to explore the predictive capability as all 2-RT tissues were from the complete response condition.


This review presents a birds-eye view of the recent developments in optical spectroscopy in the diagnosis of cervical cancers. These techniques have been shown to provide noninvasive/less invasive means for the diagnosis of cervical caners. In the case of LIF, multi-digital colposcope is being evaluated in clinical conditions. FTIR spectroscopy demonstrates the ability to diagnose cervix cancer using exfoliated cells and formalin-fixed paraffin embedded tissues. However, for in vivo diagnosis, the application of this technique is very much limited due to the presence of strong water bands. But this is a very useful method for mapping/imaging as well as for optical histopathology and optical diagnosis using ex vivo handled tissues. Few Raman spectroscopic studies that have been reported so far demonstrate its potential in cervical cancer diagnosis. Our Raman spectroscopy studies yielded encouraging results. Moreover, our methods, user-friendly 'limit test' methodology, can provide objective discrimination and could be practical in routine clinical practice.

Prospectively, by developing models encompassing different pathological conditions and rigorous validation of the models using multicentric blinded studies, optical diagnosis of cervical cancers can be realized. In the case of Raman spectroscopy, with the development of a suitable fiberoptic probe, implementation of the innovative technique of noninvasive in vivo Raman spectroscopic diagnosis in the clinic can become a reality.

So far, other than our own studies wherein the feasibility of Raman spectroscopic prediction of tumor radioresponse was demonstrated, [101],[102] optical spectroscopy is limited to diagnostic or screening applications. Therefore another interesting area for application of optical spectroscopy methods is prediction of therapeutic response or prognosis.


The data presented in the review was acquired from the project entitled 'Laser spectroscopy as a predictor of tumor response to radiation therapy in cervical cancer' (2003/34/17/BRNS/1903) supported by the Department of Atomic Energy; Board of Research in Nuclear Sciences, Government of India. One of the authors, Maheedhar Kodali, is thankful to DAE-BRNS for a Junior Research Fellowship. Ms. Keerti and Mr. Chetan Anand are gratefully acknowledged for their technical assistance. Dr. Sajan D. George is acknowledged for his critical comments and suggestions. We also gratefully acknowledge our senior colleague Prof. V.B. Kartha for his support and encouragement in the execution of the project.


1Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics. CA Cancer Clin 2005;55:74-108.
2Kjellberg L, Hallmans G, Mahadevan-Jansen A, Johansson R, Bergman F, Wodell G, et al . Smoking, diet, pregnancy and oral contraceptives use as cervical intra epithelial neoplasia in relation to Human Papilloma virus infection. Br J Cancer 2000;82:1332-8.
3Notani PN. Global variation in cancer incidence and mortality. Curr Sci 2001;81:465-74.
4Mitchell MF. Accuracy of colposcopy. Consult Obstet Gynecol 1994;6:70-3.
5Koss LG. The Papanicolaou test for cervical cancer detection: A triumph and a tragedy. J Am Med Assoc 1989;261:737-45.
6Jadhav GK, Bhanumathi P, Uma Devi P, Seetharamaiah T, Vidyasagar MS, Koteshwar Rao K, et al . Possible role of glutathione in predicting radiotherapy response of cervix cancer. Int J Radiat Oncol Biol Phys 1998;41:3-5.
7Russo A, Mitchell JB, Finkelstein E, DeGraff WG, Spiro IJ, Gamson J. The effects of cellular glutathione elevation on the oxygen enhancement ratio. Radiat Res 1985;103:232-9.
8Bhattathiri VN, Sreelekha TT, Sebastian P, Remani P, Chandini R, Vijayakumar T, et al . Influence of plasma GSH level on acute radiation mucositis of the oral cavity. Int J Radiat Oncol Biol Phys 1994;29:383-6.
9World Health Organization. WHO handbook for reporting results of cancer treatment. WHO offset publication no. 48. World Health Organization: Geneva; 1979.
10Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, et al . New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst 2000;92:205-16.
11Ohara K, Oki A, Tanaka YO, Onishi K, Fukumitsu N, Hashimoto T, et al . Early determination of uterine cervical squamous cell carcinoma radioresponse identifies high- and low-response tumors. Int J Radiat Oncol Biol Phys 2006;64:1179-82.
12Nakano T, Oka K, Ishikawa A, Morita S. Immunohistochemical prediction of radiation response and local control in radiation therapy for cervical cancer. Cancer Detect Prev 1998;22:120-8.
13Bolger BS, Symonds RP, Stanton PD, MacLean AB, Burnett R, Kelly P, et al . Prediction of radiotherapy response of cervical carcinoma through measurement of proliferation rate. Br J Cancer 1996;74:1223-6.
14Rofstad EK. NMR spectroscopy in prediction and monitoring of radiation response of tumors in vivo. Int J Radiat Biol 1990;57:1-5.
15Kortum RR. Quantitative optical spectroscopy for tissue diagnosis. Ann Rev Phys Chem 1996;47:555-606.
16Wagnieres GA, Star WM, Wilson BC. In vivo fluorescence spectroscopy and imaging for oncological applications. Photochem Photobiol 1998;68:603-32.
17Haringsma J, Tytgat GN. Fluorescence and autofluorescence. Baillieres Best Pract Res Clin Gastroenterol 1999;13:1-10.
18Ramanujam N. Fluorescence spectroscopy of neoplastic and non-neoplastic tissues. Neoplasia 2000;2:89-117.
19Sokolov K, Follen M, Kortum RR. Optical spectroscopy for detection of neoplasia. Curr Opin Chem Biol 2002;6:651-8.
20Mayinger B. Endoscopic fluorescence spectroscopic imaging in the gastrointestinal tract. Gastrointest Endosc Clin N Am 2004;14:487-505.
21De Veld DC, Witjes MJ, Sterenborg HJ, Roodenburg JL. The status of in vivo autofluorescence spectroscopy and imaging for oral oncology. Oral Oncol 2005;41:117-31.
22Kartha VB, Kurien J, Pai KM, Rao L, Rai L, Mahato KK, et al . Diagnosis at the molecular level:Analytical laser spectroscopy for clinical applications, Photo/Electrochemistry and Photobiology in the Environment, Energy and Fuel, 2005;153-221.
23Lacroix M, Poinsot V, Fournier C, Couderc F. Laser-induced fluorescence detection schemes for the analysis of proteins and peptides using capillary electrophoresis. Electrophoresis 2005;26:2608-21.
24Aaron JJ, Trajkovska S. Fluorescence studies of anti-cancer drugs-analytical and biomedical applications. Curr Drug Targets 2006;7:1067-81.
25Markovich RJ, Pidgeon C. Introduction to Fourier transform infrared spectroscopy and applications in the pharmaceutical sciences. Pharm Res 1991;8:663-75.
26Franck P, Nabet P, Dousset B. Applications of infrared spectroscopy to medical biology. Cell Mol Biol (Noisy-le-grand) 1998;44:273-5.
27Sahu RK, Mordechai S. Fourier Transform infrared spectroscopy in cancer detection. Future Oncol 2005;1:635-7.
28Murali Krishna C, Kegelaer G, Adt I, Rubin S, Kartha VB, Manfait M, et al . Combined fourier transform infrared and Raman spectroscopic approach for identification of multidrug resistance phenotype in cancer cell lines. Biopolymers 2006;82:462-70.
29Bonnier F, Rubin S, Venteo L, Murali Krishna C, Pluot M, Baehrel B, et al . In-vitro analysis of normal and aneurismal human ascending aortic tissues using FTIR microspectroscopy. Biochim Biophys Acta 2006;1758:968-73.
30Murali Krishna C, Sockalingum GD, Bhat RA, Venteo L, Kushtagi P, Pluot M, et al . FTIR and Raman microspectroscopy of normal, benign and malignant formalin-fixed ovarian tissues. Anal Bioanal Chem 2007;387:1649-56.
31Hanlon EB, Manoharan R, Koo TW, Shafer KE, Motz JT, Fitzmaurice M, et al . Prospects for in vivo Raman spectroscopy. Phys Med Biol 2000;45:R1-59.
32Choo-Smith LP, Edwards HG, Endtz HP, Kros JM, Heule F, Barr H, et al . Medical applications of Raman spectroscopy: From proof of principle to clinical implementation. Biopolymers 2002;67:1-9.
33Murali Krishna C, Sockalingum GD, Kurien J, Rao L, Venteo L, Pluot M, et al . Micro-Raman spectroscopy for optical pathology of oral squamous cell carcinoma. Appl Spectrosc 2004;58:1128-35.
34Murali Krishna C, Sockalingum GD, Venteo L, Bhat RA, Kushtagi P, Pluot M, et al . Evaluation of the suitability of ex vivo handled ovarian tissues for optical diagnosis by Raman microspectroscopy. Biopolymers 2005;79:269-76.
35Krafft C, Sergo V. Biomedical applications of Raman and infrared spectroscopy to diagnose tissues. Spectroscopy 2006;20:195-218.
36Lin SY, Li MJ, Cheng WT. FTIR and Raman vibrational microspectroscopies used for spectral biodiagnosis of human tissues. Spectroscopy 2007;21:1-30.
37Alfano RR, Tata DB, Cordero J, Tomashefsky P, Longo FW, Alfano MA. Laser induced flourescence spectroscopy from native cancerous and normal tissue. IEEE J Quant Elect 1984;20:1507-11.
38Mahadevan- Jasen A, Mitchell MF, Silva E, Thomsen S, Kortum RR. Study of the fluorescence properties of normal and neoplastic human cervical tissue. Lasers Surg Med 1993;13:647-55.
39Ramanujam N, Mitchell MF, Mahadevan-Jasen A, Thomsen S, Silva E, Kortum RR. Fluorescence spectroscopy: A diagnostic tool for cervical intraepithelial neoplasia (CIN). Gynecol Oncol 1994;52:31-8.
40Ramanujam N, Mitchell MF, Mahadevan-Jasen A, Warren S, Thomsen S, Silva E, et al . In vivo diagnosis of cervical intraepithelial neoplasia using 337-nm-excited laser-induced fluorescence. Proc Natl Acad Sci USA 1994;91:10193-7.
41Atkinson EN, Mitchell MF, Ramanujam N, Kortum RR. Statistical techniques for diagnosing CIN using fluorescence spectroscopy: SVD and CART. J Cell Biochem Suppl 1995;23:125-30.
42Ramanujam N, Mitchell MF, Mahadevan-Jasen A, Thomsen S, Malpica A, Wright T, et al . Development of a multivariate statistical algorithm to analyze human cervical tissue fluorescence spectra acquired in vivo . Lasers Surg Med 1996;19:46-62.
43Mitchell MF, Cantor SB, Brookner C, Utzinger U, Schottenfeld D, Kortum RR. Screening for squamous intraepithelial lesions with fluorescence spectroscopy. Obstet Gynecol 1999;94:889-96.
44Mitchell MF, Cantor SB, Ramanujam N, Tortolero-Luna G, Kortum RR. Fluorescence spectroscopy for diagnosis of squamous intraepithelial lesions of the cervix. Obstet Gynecol 1999;93:462-70.
45Drezek R, Sokolov K, Utzinger U, Boiko I, Malpica A, Follen M, et al . Understanding the contributions of NADH and collagen to cervical tissue fluorescence spectra: Modeling, measurements and implications. J Biomed Opt 2001;6:385-96.
46Georgakoudi I, Jacobson BC, Müller MG, Sheets EE, Badizadegan K, Carr-Locke DL, et al . NAD(P)H and collagen as in vivo quantitative fluorescent biomarkers of epithelial precancerous changes. Cancer Res 2002;62:682-7.
47Chang SK, Dawood MY, Staerkel G, Utzinger U, Atkinson EN, Kortum RR, et al . Fluorescence spectroscopy for cervical precancer detection:Is there variance across the menstrual cycle? J Biomed Opt 2002;7:595-602.
48Cox DD, Chang SK, Dawood MY, Staerkel G, Utzinger U, Kortum RR, et al . Detecting the signal of the menstrual cycle in fluorescence spectroscopy of the cervix. Appl Spectrosc 2003;57:67-72.
49Brookner C, Utzinger U, Follen M, Kortum RR, Cox D, Atkinson EN. Effects of biographical variables on cervical fluorescence emission spectra. J Biomed Opt 2003;8:479-83.
50Freeberg JA, Serachitopol DM, McKinnon N, Price R, Atkinson EN, Cox DD, et al . Fluorescence and reflectance device variability throughout the progression of a phase II clinical trial to detect and screen for cervical neoplasia using a fiber optic probe. J Biomed Opt 2007;12:034015.
51Chang SK, Mirabal YN, Atkinson EN, Cox D, Malpica A, Follen M, et al . Combined reflectance and fluorescence spectroscopy for in vivo detection of cervical pre-cancer. J Biomed Opt 2005;10:024031.
52Lee JS, Shuhatovich O, Price R, Pikkula B, Follen M, McKinnon N, et al . Design and preliminary analysis of a study to assess intra-device and inter-device variability of fluorescence spectroscopy instruments for detecting cervical neoplasia. Gynecol Oncol 2005;99:S98-S111.
53Pikkula BM, Shuhatovich O, Price RL, Serachitopol DM, Follen M, McKinnon N, et al . Instrumentation as a source of variability in the application of fluorescence spectroscopic devices for detecting cervical neoplasia. Biomed Opt 2007;12:034014.
54Milbourne A, Park SY, Benedet JL, Miller D, Ehlen T, Rhodes H, et al . Results of a pilot study of multispectral digital colposcopy for the in vivo detection of cervical intraepithelial neoplasia. Gynecol Oncol 2005;99:S67-75.
55Georgakoudi I, Sheets EE, Müller MG, Backman V, Crum CP, Badizadegan K, et al . Trimodal spectroscopy for the detection and characterization of cervical precancers in vivo. Cancer Res 2002;62:682-7.
56Schomacker KT, Meese TM, Jiang C, Abele CC, Dickson K, Sum ST, et al . Novel optical detection system for in vivo identification and localization of cervical intraepithelial neoplasia. J Biomed Opt 2006;11:34009.
57Majumder SK, Uppal A, Gupta PK. In-vitro diagnosis of human uterine malignancy using N2 laser induced autofluorescence spectroscopy. Curr Sci 1996;70:833-6.
58Chidananda SM, Satyamoorthy K, Rai L, Manjunath AP, Kartha VB. Optical diagnosis of cervical cancer by fluorescence spectroscopy technique. Int J Cancer 2006;119:139-45.
59Palsson S, Stenram U, Thompson MS, Vaitkuviene A, Poskiene V, Ziobakiene R, et al . Methods for detailed histopathological investigation and localization of biopsies from cervix uteri to improve the interpretation of autofluorescence data. J Environ Pathol Toxicol Oncol 2006;25:321-40.
60DeSantis T, Chakhtoura N, Twiggs L, Ferris D, Lashgari M, Flowers L, et al . Spectroscopic imaging as a triage test for cervical disease:a prospective multicenter clinical trial. J Low Genit Tract Dis 2007;11:18-24.
61Wu Y, Zheng W, Qu JY. Sensing cell metabolism by time-resolved autofluorescence. Opt Lett 2006;31:3122-4.
62Alvarez RD, Wright TC; Optical Detection Group. Effective cervical neoplasia detection with a novel optical detection system: A randomized trial. Gynecol Oncol 2007;104:281-9.
63Martin SF, Wood AD, McRobbie MM, Mazilu M, McDonald MP, Samuel ID, et al . Fluorescence spectroscopy of an in vitro model of human cervical precancer identifies neoplastic phenotype. Int J Cancer 2007;120:1964-70.
64Werner CL, Griffith WF, Ashfaq R, Gossett D, Wilkinson E, Raab S, et al . Comparison of human papilloma virus testing and spectroscopy combined with cervical cytology for the detection of high-grade cervical neoplasia. J Low Genit Tract Dis 2007;11:73-9.
65Wong PT, Wong K, Fung Kee Fung M. Pressure-tuning FTIR study of human cervical tissues. Appl Spectroscopy 1993;47:1058-63.
66Yazdi HM, Bertrand MA, Wong PT. Detecting structural changes at the molecular level with Fourier transform infrared spectroscopy: A potential tool for prescreening preinvasive lesions of the cervix. Acta Cytol 1996;40:664-8.
67Morris BJ, Lee C, Nightingale BN, Molodysky E, Morris LJ, Appio R, et al . Fourier transform infrared spectroscopy of dysplastic, papillomavirus-positive cervicovaginal lavage specimens. Gynecol Oncol 1995;56:245-9.
68Fung Kee Fung M, Senterman M, Eid P, Faught W, Mikhael NZ, Wong PT. Comparison of Fourier-transform infrared spectroscopic screening of exfoliated cervical cells with standard Papanicolaou screening. Gynecol Oncol 1997;66:10-5.
69Cohenford MA, Godwin TA, Cahn F, Bhandare P, Caputo TA, Rigas B. Infrared spectroscopy of normal and abnormal cervical smears: Evaluation by principal component analysis. Gynecol Oncol 1997;66:59-65.
70Sindhuphak R, Issaravanich S, Udomprasertgul V, Srisookho P, Warakamin S, Sindhuphak S, et al . A new approach for the detection of cervical cancer in Thai women. Gynecol Oncol 2003;90:10-4.
71Chiriboga L, Xie P, Yee H, Zarou D, Zakim D, Diem M. Infrared spectroscopy of human tissue. IV. Detection of dysplastic and neoplastic changes of human cervical tissue via infrared microscopy. Cell Mol Biol (Noisy-le-grand) 1998;44:219-29.
72Miljkovic M, Romeo M, Matthδus C, Diem M. Infrared microspectroscopy of individual human cervical cancer (HeLa) cells suspended in growth medium. Biopolymers 2004;74:172-5.
73Romeo M, Matthδus C, Miljkovic M, Diem M. Infrared microspectroscopy of individual human cervical cancer (HeLa) cells. Biopolymers 2004;74:168-71.
74Boydston-White S, Romeo M, Chernenko T, Regina A, Miljkoviζ M, Diem M. Cell-cycle-dependent variations in FTIR micro-spectra of single proliferating HeLa cells: Principal component and artificial neural network analysis. Biochim Biophys Acta 2006;1758:908-14.
75Diem M, Romeo M, Boydston-White S, Miljkovic M, Matthaus C. A decade of vibrational micro-spectroscopy of human cells and tissue (1994-2004). Analyst 2004;129:880-5.
76Chiriboga L, Yee H, Diem M, Wood B. Infrared spectral features of exfoliated cervical cells, cervical adenocarcinoma tissue and an adenocarcinoma cell line (SiSo). Gynecol Oncol 2002;85:170-4.
77Wood BR, Chiriboga L, Yee H, Quinn MA, McNaughton D, Diem M. Fourier transform infrared (FTIR) spectral mapping of the cervical transformation zone and dysplastic squamous epithelium. Gynecol Oncol 2004;93:59-68.
78Mourant JR, Gibson RR, Johnson TM, Carpenter S, Short KW, Yamada YR, et al . Methods for measuring the infrared spectra of biological cells. Phys Med Biol 2003;48:243-57.
79Wood BR, Quinn MA, Tait B, Ashdown M, Hislop T, Romeo M, et al . FTIR microspectroscopic study of cell types and potential confounding variables in screening for cervical malignancies. Biospectroscopy 1998;4:75-91.
80Romeo MJ, Wood BR, Quinn MA, McNaughton D. Removal of blood components from cervical smears: Implications for cancer diagnosis using FTIR spectroscopy. Biopolymers 2003;72:69-76.
81Shaw RA, Guijon FB, Paraskevas M, Ying SL, Mantsch HH. Infrared spectroscopy of exfoliated cervical cell specimens: Proceed with caution. Anal Quant Cytol Histol 1999;21:292-302.
82Chang JI, Huang YB, Wu PC, Chen CC, Huang SC, Tsai YH. Characterization of human cervical precancerous tissue through the fourier transform infrared microscopy with mapping method. Gynecol Oncol 2003;91:577-83.
83Mark S, Sahu RK, Kantarovich K, Podshyvalov A, Guterman H, Goldstein J, et al . Fourier transform infrared microspectroscopy as a quantitative diagnostic tool for assignment of premalignancy grading in cervical neoplasia. J Biomed Opt 2004;9:558-67.
84Mordechai S, Sahu RK, Hammody Z, Mark S, Kantarovich K, Guterman H, et al . Possible common biomarkers from FTIR microspectroscopy of cervical cancer and melanoma. J Microsc 2004;215:86-91.
85Podshyvalov A, Sahu RK, Mark S, Kantarovich K, Guterman H, Goldstein J, et al . Distinction of cervical cancer biopsies by use of infrared microspectroscopy and probabilistic neural networks. Appl Opt 2005;44:3725-34.
86Walsh MJ, Singh MN, Pollock HM, Cooper LJ, German MJ, Stringfellow HF, et al . ATR microspectroscopy with multivariate analysis segregates grades of exfoliative cervical cytology. Biochem Biophys Res Commun 2007;352:213-9.
87Mahadevan-Jansen A, Mitchell MF, Ramanujam N, Malpica A, Thomsen S, Utzinger U, et al . Near-infrared Raman spectroscopy for in vitro detection of cervical precancers. Photochem Photobiol 1998;68:123-32.
88Mahadevan-Jansen A, Mitchell MF, Ramanujam N, Utzinger U, Kortum RR. Development of a fiber optic probe to measure NIR Raman spectra of cervical tissue in vivo . Photochem Photobiol 1998;68:427-31.
89Utzinger UR, Heintzelman DL, Mahadevan-Jansen A, Malpica A, Michele Follen, Kortum RR. Near-infrared Raman spectroscopy for in vivo detection of cervical precancers. Appl Spectroscopy, 1998;55:955-9.
90Lyng FM, Faolαin EO, Conroy J, Meade AD, Knief P, Duffy B, et al . Vibrational spectroscopy for cervical cancer pathology, from biochemical analysis to diagnostic tool. Exp Mol Pathol 2007;82:121-9.
91Murali Krishna C, Prathima NB, Malini R, Vadhiraja BM, Bhat RA, Donald JF, et al . Raman spectroscopy studies for diagnosis of cancers in human uterine cervix. Vib Spectrosc 2006;41:136-41.
92Parker ES. Applications of infrared and Raman and resonance Raman spectroscopy in biochemistry. Plenum Press: New York; 1983.
93Tonge PJ, Carey PR, Raman, resonance raman and FTIR spectroscopic studies of enzyme-substrate complexes. In : Clark RJ, Hester RE, editors. Biomolecular Spectroscopy Part A, Advances in Spectroscopy, Vol 20, John Wiley and Sons: Chchester; p. 129 (Chapter 3).
94Mahalanobis JC. On generalized distance in statistics. Proc Natl Inst Sci India 1936;2:49-55.
95Mark HL. Use of mahalanobis distances to evaluate sample preparation methods for near-infrared reflectance analysis. Anal Chem 1987;59:790-5.
96PLSplus/IQ User's Guide. Galactic Industries Corporation.
97Malini R, Venkatakrishna K, Kurien J, Pai KM, Lakshmi Rao, Kartha VB, et al . Discrimination of normal, inflammatory, premalignant and malignant oral tissue: A Raman spectroscopy study. Biopolymers 2006;81:179-93.
98Chowdary MV, Kalyan KK, Kurien J, Mathew S, Murali Krishna C. Discrimination of normal, benign and malignant breast tissue by Raman Spectroscopy. Biopolymers 2006;83:556-69.
99Kalyan KK, Anand A, Chowdary MV, Keerthi, Kurien J, Murali Krishna C, et al . Discrimination of normal and malignant stomach mucosal tissues by Raman spectroscopy: A pilot study. Vib Spectrosc 2007;44:382-7.
100Chowdary MV, Kalyan KK, Keerthi, Anand A, Kurien J, Murali Krishna C, et al . Discrimination of normal and malignant mucosal tissue of colon by Raman spectroscopy. Photomed Laser Surg 2007;25:269-74.
101Vidyasagar MS, Maheedhar K, Vadiraja BM, Donald JF, Kartha VB, Murali Krishna C. Prediction of radiotherapy response in cervix cancer by Raman spectroscopy: A pilot study. Biopolymers 2008, In Press.
102Vidyasagar MS, Maheedhar K, Vadiraja BM, Donald JF, Kartha VB, Murali Krishna C. Raman spectroscopy of tissues collected at different fractions of radiation therapy: Response assessment to radiotherapy in cervix cancers. Int J Radiat Oncol Biol Physics 2007;69:S388-9.
103Murali Krishna C, Sockalingum GD, Vadhiraja BM, Maheedhar K, Anuradha Rao CK, Rao L, et al . Vibrational spectroscopic study of formalin fixed cervix tissues. Biopolymers 2007;85:214-21.