Quantification of metabolites in breast cancer patients with different clinical prognosis using HR MAS MR spectroscopy

Absolute quantitative measures of breast cancer tissue metabolites can increase our understanding of biological processes. Electronic REference To access In vivo Concentrations (ERETIC) was applied to high resolution magic angle spinning MR spectroscopy (HR MAS MRS) to quantify metabolites in intact breast cancer samples. The ERETIC signal was calibrated using solutions of creatine and TSP. The largest relative errors of the ERETIC method were 8.4%, compared to 4.4% for the HR MAS MRS method using TSP as a standard. The same MR experimental procedure was applied to intact tissue samples from breast cancer patients with clinically defined good (n = 13) and poor (n = 16) prognosis. All samples were examined by histopathology for relative content of different tissue types and proliferation index (MIB‐1) after MR analysis. The resulting spectra were analyzed by quantification of tissue metabolites (β‐glucose, lactate, glycine, myo‐inositol, taurine, glycerophosphocholine, phosphocholine, choline and creatine), by peak area ratios and by principal component analysis. We found a trend toward lower concentrations of glycine in patients with good prognosis (1.1 µmol/g) compared to patients with poor prognosis (1.9 µmol/g, p = 0.067). Tissue metabolite concentrations (except for β‐glucose) were also found to correlate to the fraction of tumor, connective, fat or glandular tissue by Pearson correlation analysis. Tissue concentrations of β‐glucose correlated to proliferation index (MIB‐1) with a negative correlation factor (?0.45, p = 0.015), consistent with increased energy demand in proliferating tumor cells. By analyzing several metabolites simultaneously, either in ratios or by metabolic profiles analyzed by PCA, we found that tissue metabolites correlate to patients' prognoses and health status five years after surgery. This study shows that the diagnostic and prognostic potential in MR metabolite analysis of breast cancer tissue is greater when combining multiple metabolites (MR Metabolomics). Copyright © 2010 John Wiley & Sons, Ltd.     

High-resolution magic angle spinning MRS of breast cancer tissue

High-resolution magic angle spinning (HR MAS) may develop into a new diagnostic tool for studying intact tissue samples, and several types of cancer have been investigated with promising results. In this study HR MAS spectra of breast cancer tissue from 10 patients have been compared to conventional high-resolution spectra of perchloric acid extracts of the same tissue type. The HR MAS spectra show resolution comparable to spectra of extracts, and two-dimensional techniques lead to identification of a majority of the constituents. More than 30 different metabolites have been detected and assigned. To our knowledge this is the most detailed assignment of biochemical components in intact human breast tissue. The spectra of intact breast cancer tissue differ from perchloric acid extracts by the presence of lipids and fewer signals in the low field region. HR MAS analysis of intact breast tissue specimens is a rapid method, providing spectra with resolution where relative quantification of the majority of the detected metabolites is possible.     

1H HR‐MAS and genomic analysis of human tumor biopsies discriminate between high and low grade astrocytomas

We investigate the profile of choline metabolites and the expression of the genes of the Kennedy pathway in biopsies of human gliomas (n = 23) using ¹H High Resolution Magic Angle Spinning (HR‐MAS, 11.7 Tesla, 277 K, 4000 Hz) and individual genetic assays. ¹H HR‐MAS spectra allowed the resolution and relative quantification by the LCModel of the resonances from choline (Cho), phosphocholine (PC) and glycerophosphorylcholine (GPC), the three main components of the combined tCho peak observed in gliomas by in vivo ¹H NMR spectroscopy. All glioma biopsies depicted a prominent tCho peak. However, the relative contributions of Cho, PC, and GPC to tCho were different for low and high grade gliomas. Whereas GPC is the main component in low grade gliomas, the high grade gliomas show a dominant contribution of PC. This circumstance allowed the discrimination of high and low grade gliomas by ¹H HR‐MAS, a result that could not be obtained using the tCho/Cr ratio commonly used by in vivo ¹H NMR spectroscopy. The expression of the genes involved in choline metabolism has been investigated in the same biopsies. High grade gliomas depict an upregulation of the β gene of choline kinase and phospholipase C, as well as a downregulation of the cytidyltransferase B gene, the balance of these being consistent with the accumulation of PC. In the low grade gliomas, phospholipase A₁ and lysophospholypase are upregulated and phospholipase D is downregulated, supporting the accumulation of GPC. The present findings offer a promising procedure that will potentially help to accurately grade glioma tumors using ¹H HR‐MAS, providing in addition the genetic background for the alterations of choline metabolism observed in high and low grade gliomas. Copyright © 2009 John Wiley & Sons, Ltd.     

Quantification of phosphocholine and glycerophosphocholine with 31P edited 1H NMR spectroscopy

Choline and the related compounds phosphocholine (PC) and glycerophosphocholine (GPC) are considered to be important metabolites in oncology. Past studies have demonstrated correlations linking the relative ratios and concentrations of these metabolites with the development and progression of cancer. Currently, in vivo and tissue ex vivo magnetic resonance spectroscopy methods have mostly centered on measuring the total concentration of these metabolites and have difficulty in differentiating between them. Here, a new scheme that uses (31)P edited (1)H spectroscopy to quantify the concentrations of choline, PC and GPC in biological samples is reported and its applicability is demonstrated using samples of human brain tumor extracts. This method is particularly well-suited for analytical situations where the PC and GPC resonances are not sufficiently resolved and/or are obscured by other metabolites. Consequently, this scheme has the potential to be used for the analysis of choline compounds in ex vivo tissue samples.     

Assessment of early docetaxel response in an experimental model of human breast cancer using DCE‐MRI,ex vivo HR MAS,and in vivo 1H MRS

The purpose of this study was to evaluate the use of dynamic contrast‐enhanced (DCE) MRI, in vivo ¹H MRS and ex vivo high resolution magic angle spinning (HR MAS) MRS of tissue samples as methods to detect early treatment effects of docetaxel in a breast cancer xenograft model (MCF‐7) in mice. MCF‐7 cells were implanted subcutaneously in athymic mice and treated with docetaxel (20, 30, and 40 mg/kg) or saline six weeks later. DCE‐MRI and in vivo ¹H MRS were performed on a 7 T MR system three days after treatment. The dynamic images were used as input for a two‐compartment model, yielding the vascular parameters K^trans and v_e. HR MAS MRS, histology, and immunohistochemical staining for proliferation (Ki‐67), apoptosis (M30 cytodeath), and vascular/endothelial cells (CD31) were performed on excised tumor tissue. Both in vivo spectra and HR MAS spectra were used as input for multivariate analysis (principal component analysis (PCA) and partial least squares regression analysis (PLS)) to compare controls to treated tumors. Tumor growth was suppressed in docetaxel‐treated mice compared to the controls. The anti‐tumor effect led to an increase in K^trans and v_e values in all the treated groups. Furthermore, in vivo MRS and HR MAS MRS revealed a significant decrease in choline metabolite levels for the treated groups, in accordance with reduced proliferative index as seen on Ki‐67 stained sections. In this study DCE‐MRI, in vivo MRS and ex vivo HR MAS MRS have been used to demonstrate that docetaxel treatment of a human breast cancer xenograft model results in changes in the vascular dynamics and metabolic profile of the tumors. This indicates that these MR methods could be used to monitor intra‐tumoral treatment effects. Copyright © 2009 John Wiley & Sons, Ltd.     

Preinvasive and invasive cervical cancer: an ex vivo proton magic angle spinning magnetic resonance spectroscopy study

The aim of this study was to obtain (1)H MR spectra using magic angle spinning (MAS) techniques from punch biopsies (<20 mg) of preinvasive and invasive cervical disease and to correlate the spectral profiles with sample classification on the basis of histopathology. Tissue samples were obtained at colposcopic examination, during local treatment of cervical intraepithelial neoplasia (CIN) or at hysterectomy. (1)H MAS MRS was performed at 25 degrees C while spinning the sample at 4.5 kHz. After measurement, the tissue was immersed in formalin and the pathology determined. Histological examination after (1)H MAS MRS defined 27 samples with squamous cell carcinoma (SCC), 12 with CIN and 39 with only normal tissue. The standardized integrals of the lipid, choline and creatine regions of the spectra were significantly higher in SCC than in normal or CIN tissue. There was no obvious difference in the standardized integral of the region 4.15-3.5 ppm. The acyl fatty acid side-chain length was longer or less unsaturated in SCC than in normal tissue. Normal tissue from patients with SCC showed significantly higher triglycerides than normal tissue from patients with benign uterine disease but significantly lower triglycerides than SCC tissue. (1)H MAS MRS of the uterine cervix ex vivo may be used to differentiate non-invasive from invasive cervical lesions, increase interpretation of in vivo MRS and provide insights into tumor biology.     

Evaluation of 31P high-resolution magic angle spinning of intact tissue samples

The first detailed evaluation is presented of high-resolution (31)P MRS using magic angle spinning (MAS) of intact tissue samples and comparison with the conventional method of studying tissue extracts. The main motivation is that MAS leaves the sample intact at the end of the study for histopathological evaluation. While MAS of tissue samples has previously been demonstrated for (1)H MRS, (31)P MRS is better suited to study of the phospholipid metabolites of importance in cancer. Samples of rhabdomyosarcoma and RIF-1 experimental tumours were maintained at 4 degrees C, spun at 3 kHz and measured in 28-min acquisitions at 11.7 and 14 T. Metabolite stability was evaluated using four sequential 28-min acquisitions. High-resolution MRS was performed on extracts of the same tissue samples. (31)P HR-MAS yielded well-resolved high-resolution spectra, showing peaks from phosphoethanolamine (PE), phosphocholine (PC), inorganic phosphate, glycerophosphoethanolamine and glycerophosphocholine, with linewidths in the range 3-20 Hz. In tumour samples there was no significant change in peak areas over a 2-h period, while peaks sensitive to pH (inorganic phosphate, PE and PC) showed a small change in chemical shift, corresponding to a change of 0.13 +/- 0.06 pH units. Tissue metabolite concentrations showed good agreement with concentrations measured from extracts of the same pieces of tissue. For calculation of metabolite concentrations, the measurement of a reference compound in a separate measurement is more robust than using the signal from a reference compound in the rotor with the sample. Compared with performing tissue extracts, use of MAS of intact tissue samples requires less preparation, is quicker and permits the same sample to be used for subsequent histopathology. The methodology has particular application in studying phospholipid metabolism in cancer and in monitoring tumour response to treatment, where concentrations of phospholipid-related metabolites are found to alter following response to a wide range of anti-cancer therapies.     

The PCA and LDA analysis on the differential expression of proteins in breast cancer

Breast cancer is a leading cause of mortality in women. In Malaysia, it is the most common cancer to affect women. The most common form of breast cancer is infiltrating ductal carcinoma (IDC). A proteomic approach was undertaken to identify protein profile changes between cancerous and normal breast tissues from 18 patients. Two protein extracts; aqueous soluble and membrane associated protein extracts were studied. Thirty four differentially expressed proteins were identified. The intensities of the proteins were used as variables in PCA and reduced data of six principal components (PC) were subjected to LDA in order to evaluate the potential of these proteins as collective biomarkers for breast cancer. The protein intensities of SEC13-like 1 (isoform b) and calreticulin contributed the most to the first PC while the protein intensities of fibrinogen beta chain precursor and ATP synthase D chain contributed the most to the second PC. Transthyretin precursor and apolipoprotein A-1 precursor contributed the most to the third PC. The results of LDA indicated good classification of samples into normal and cancerous types when the first 6 PCs were used as the variables. The percentage of correct classification was 91.7% for the originally grouped tissue samples and 88.9% for cross-validated samples.     

Silencing of the glycerophosphocholine phosphodiesterase GDPD5 alters the phospholipid metabolite profile in a breast cancer model in vivo as monitored by 31P MRS

Abnormal choline phospholipid metabolism is an emerging hallmark of cancer, which is implicated in carcinogenesis and tumor progression. The malignant metabolic phenotype is characterized by high levels of phosphocholine (PC) and relatively low levels of glycerophosphocholine (GPC) in aggressive breast cancer cells. Phosphorus (³¹P) MRS is able to non‐invasively detect these water‐soluble metabolites of choline as well as ethanolamine phospholipid metabolism. Here we have investigated the effects of stably silencing glycerophosphoester diesterase domain containing 5 (GDPD5), which is an enzyme with glycerophosphocholine phosphodiesterase activity, in MDA‐MB‐231 breast cancer cells and orthotopic tumor xenografts. Tumors in which GDPD5 was stably silenced with GDPD5‐specific shRNA contained increased levels of GPC and phosphoethanolamine (PE) compared with control tumors. Copyright © 2014 John Wiley & Sons, Ltd.     

Correlation of phospholipid metabolites with prostate cancer pathologic grade, proliferative status and surgical stage - impact of tissue environment

This study investigates the relationship between phospholipid metabolite concentrations, Gleason score, rate of cellular proliferation and surgical stage in malignant prostatectomy samples by performing one- and two-dimensional, high-resolution magic angle spinning, total correlation spectroscopy, pathology and Ki-67 staining on the same surgical samples. At radical prostatectomy, surgical samples were obtained from 49 patients [41 with localized TNM stage T1 and T2, and eight with local cancer spread (TNM stage T3)]. Thirteen of the tissue samples were high-grade prostate cancer [Gleason score: 4?+?3 (n?=?7); 4?+?4 (n?=?6)], 22 low-grade prostate cancer [Gleason score: 3?+?3 (n?=?17); 3?+?4 (n?=?5)] and 14 benign prostate tissues. This study demonstrates that high-grade prostate cancer shows significantly higher Ki-67 staining and concentrations of phosphocholine (PC) and glycerophosphocholine (GPC) than does low-grade prostate cancer (2.4?±?2.8% versus 7.6?±?3.5%, p?相似文献   

Evidence supporting the association of polyomavirus BK genome with prostate cancer

Prostate cancer (PCA) is the most frequent cancer in men. Exposure to infectious agents has been reported to have a putative role in tumorigenesis. Among the infectious agents, convincing evidence has been accumulated about the human polyomavirus BK (BKV). Tissue fresh specimens, serum, and urine samples were collected from 124 consecutive patients, 56 with PCA and 68 with benign prostatic hyperplasia (BPH). Quantitative PCR assays were used to assess the presence of BKV and JC virus (JCV) genomes. BKV-positive tissue specimens were found in 32.1 and 22.1 % of PCA and BPH patients, respectively; in PCA group the number of positive BKV specimens/patients was significantly higher than in BPH group (3.06 vs. 1.73, p = 0.02). JCV genome was found in the biopsies collected from 28.1 and 24.2 % of PCA and BPH patients, respectively, with no significant difference in the rate of JCV specimens/patients between PCA and BPH groups. Our results support the putative causal association between BKV genome and PCA. Further studies are required to demonstrate the direct pathogenetic role of BKV in the PCA occurrence and progression in order to clear the tempting way of vaccine prophylaxis.     

Detection of cancer in cervical tissue biopsies using mobile lipid resonances measured with diffusion‐weighted 1H magnetic resonance spectroscopy

The purpose of this study was to implement a diffusion‐weighted sequence for visualisation of mobile lipid resonances (MLR) using high resolution magic angle spinning (HR‐MAS) ¹H MRS and to evaluate its use in establishing differences between tissues from patients with cervical carcinoma that contain cancer from those that do not. A stimulated echo sequence with bipolar gradients was modified to allow T₁ and T₂ measurements and optimised by recording signal loss in HR‐MAS spectra as a function of gradient strength in model lipids and tissues. Diffusion coefficients, T₁ and apparent T₂ relaxation times were measured in model lipid systems. MLR profiles were characterised in relation to T₁ and apparent T₂ relaxation in human cervical cancer tissue samples. Diffusion‐weighted (DW) spectra of cervical biopsies were quantified and peak areas analysed using linear discriminant analysis (LDA). The optimised sequence reduced spectral overlap by suppressing signals originating from low molecular weight metabolites and non‐lipid contributions. Significantly improved MLR visualisation allowed visualisation of peaks at 0.9, 1.3, 1.6, 2.0, 2.3, 2.8, 4.3 and 5.3 ppm. MLR analysis of DW spectra showed at least six peaks arising from saturated and unsaturated lipids and those arising from triglycerides. Significant differences in samples containing histologically confirmed cancer were seen for peaks at 0.9 (p < 0.006), 1.3 (p < 0.04), 2.0 (p < 0.03), 2.8 (p < 0.003) and 4.3 ppm (p < 0.0002). LDA analysis of MLR peaks from DW spectra almost completely separated two clusters of cervical biopsies (cancer, ‘no‐cancer’), reflecting underlying differences in MLR composition. Generated Receiver Operating Characteristic (ROC) curves and calculated area under the curve (0.962) validated high sensitivity and specificity of the technique. Diffusion‐weighting of HR‐MAS spectroscopic sequences is a useful method for characterising MLR in cancer tissues and displays an accumulation of lipids arising during tumourigenesis and an increase in the unsaturated lipid and triglyceride peaks with respect to saturated MLR. Copyright © 2009 John Wiley & Sons, Ltd.     

Metabolomic studies of human lung carcinoma cell lines using in vitro (1)H NMR of whole cells and cellular extracts

We report principal component analysis (PCA) of (1)H NMR spectra recorded for a group of human lung carcinoma cell lines in culture and (1)H NMR analysis of extracts from the same samples. The samples studied were cells of lung tumour origin with different chemotherapy drug resistance patterns. For whole cells, it was found that the statistically significant causes of spectral variation were an increase in the choline and a decrease in the methylene mobile lipid (1)H resonance intensities, which correlate with our knowledge of the level of resistance displayed by the different cells. Similarly, in the (1)H NMR spectra of the aqueous and lipophilic extracts, significant quantitative differences in the metabolite distributions were apparent, which are consistent with the PCA results. Copyright (c) 2008 John Wiley & Sons, Ltd.     

Investigation of metabolite changes in the transition from pre-invasive to invasive cervical cancer measured using (1)H and (31)P magic angle spinning MRS of intact tissue

The aim of this study was to determine the metabolic changes in the transition from pre-invasive to invasive cervical cancer using high-resolution magic angle spinning (HR-MAS) MRS. Biopsy specimens were obtained from women with histologically normal cervix (n = 5), cervical intraepithelial neoplasia (CIN; mild, n = 5; moderate/severe, n = 40), and invasive cancer (n = 23). (1)H HR-MAS MRS data were acquired using a Bruker Avance 11.74 T spectrometer (Carr-Purcell-Meiboom-Gill sequence; TR = 4.8 s; TE = 135 ms; 512 scans; 41 min acquisition). (31)P HR-MAS spectra were obtained from the normal subjects and cancer patients only (as acetic acid applied before tissue sampling in patients with CIN impaired spectral quality) using a (1)H-decoupled pulse-acquire sequence (TR = 2.82 s; 2048 scans; 96 min acquisition). Peak assignments were based on values reported in the literature. Peak areas were measured using the AMARES algorithm. Estimated metabolite concentrations were compared between patient diagnostic categories and tissue histology using independent samples t tests. Comparisons based on patient category at diagnosis showed significantly higher estimated concentrations of choline (P = 0.0001) and phosphocholine (P = 0.002) in tissue from patients with cancer than from patients with high-grade dyskaryosis, but no differences between non-cancer groups. Division by histology of the sample also showed increases in choline (P = 0.002) and phosphocholine (P = 0.002) in cancer compared with high-grade CIN tissue. Phosphoethanolamine was increased in cancer compared with normal tissue (P = 0.0001). Estimated concentrations of alanine (P = 0.01) and creatine (P = 0.008) were significantly reduced in normal tissue from cancer patients compared with normal tissue from non-cancer patients. The estimated concentration of choline was significantly increased in CIN tissue from cancer patients compared with CIN tissue from non-cancer patients (P = 0.0001). Estimated concentrations of choline-containing metabolites increased from pre-invasive to invasive cervical cancer. Concurrent metabolite depletion occurs in normal tissue adjacent to cancer tissue.     

Spatially matched in vivo and ex vivo MR metabolic profiles of prostate cancer – investigation of a correlation with Gleason score

MR metabolic profiling of the prostate is promising as an additional diagnostic approach to separate indolent from aggressive prostate cancer. The objective of this study was to assess the relationship between the Gleason score and the metabolic biomarker (choline + creatine + spermine)/citrate (CCS/C) measured by ex vivo high‐resolution magic angle spinning MRS (HR‐MAS MRS) and in vivo MRSI, and to evaluate the correlation between in vivo‐ and ex vivo‐measured metabolite ratios from spatially matched prostate regions. Patients (n = 13) underwent in vivo MRSI prior to radical prostatectomy. A prostate tissue slice was snap‐frozen shortly after surgery and the locations of tissue samples (n = 40) collected for ex vivo HR‐MAS were matched to in vivo MRSI voxels (n = 40). In vivo MRSI was performed on a 3T clinical MR system and ex vivo HR‐MAS on a 14.1T magnet. Relative metabolite concentrations were calculated by LCModel fitting of in vivo spectra and by peak integration of ex vivo spectra. Spearman's rank correlations (ρ) between CCS/C from in vivo and ex vivo MR spectra, and with their corresponding Gleason score, were calculated. There was a strong positive correlation between the Gleason score and CCS/C measured both in vivo and ex vivo (ρ = 0.77 and ρ = 0.69, respectively; p < 0.001), and between in vivo and ex vivo metabolite ratios from spatially matched regions (ρ = 0.67, p < 0.001). Our data indicate that MR metabolic profiling is a potentially useful tool for the assessment of cancer aggressiveness. Moreover, the good correlation between in vivo‐ and ex vivo‐measured CCS/C demonstrates that our method is able to bridge MRSI and HR‐MAS molecular analysis. Copyright © 2012 John Wiley & Sons, Ltd.     

GSTM1 tissue genotype as a recurrence predictor in non-muscle invasive bladder cancer

Tissue genotyping is more useful approach than using blood genomic DNA, which can reflect the effects of the somatic mutations in cancer. Although polymorphisms in glutathione S-transferase (GST) have been associated with the risk of bladder cancer (BC) development, few reports provide information about the prognosis of BC. We investigated glutathione S-transferase mu (GSTM1) and glutathione S-transferase theta (GSTT1) genotypes using genomic DNA from primary 165 BC tissue samples to assess the association with disease prognosis. DNA samples from tumor were analyzed by multiplex polymerase chain reaction (PCR). The results were compared with clinicopathological parameters. The prognostic significance of the GSTs was evaluated by Kaplan-Meier and multivariate Cox regression model. Kaplan-Meier estimates revealed significant differences in time to tumor recurrence according to the GSTM1 tissue genotype (P = 0.038) in non-muscle invasive bladder cancer (NMIBC). Multivariate Cox regression analysis also revealed that the tissue GSTM1 genotype (hazards ratio [HR]: 0.377, P = 0.031) was an independent predictor of bladder tumor recurrence in NMIBC. This identification of GSTM1 tissue genotype as a prognosticator for determining recurrence in NMIBC should prove highly useful in a clinical setting.     

Impact of intratumoral heterogeneity on the metabolic profiling of breast cancer tissue using high-resolution magic angle spinning magnetic resonance spectroscopy

High-resolution magic angle spinning (HR-MAS) magnetic resonance spectroscopy (MRS) is a useful metabolic profiling technique for human tissue. However, the impact of intratumoral heterogeneity on the metabolite levels of breast cancers is not yet established. The purpose of this prospective study was to investigate whether the tumor cell fraction of core needle biopsy (CNB) specimens of breast cancers affect metabolic profiles assessed with HR-MAS MRS. From June 2015 to December 2016, 46 patients with 47 breast cancers were enrolled. HR-MAS MRS was used for the metabolic profiling of 285 CNB specimens from the 47 cancers. Multiple CNB samples (range 2–8) for the HR-MAS MRS experiment were obtained from surgical specimens under ultrasound guidance following surgical removal of the tumor. Tumor cell fraction was expressed as a percentage of the tumor cell volume relative to the total tumor volume contained in each CNB sample. Metabolite quantification levels were compared according to primary tumor characteristics using the t-test. Multivariate analyses were performed including primary tumor characteristics and tumor cell percentages as variables. Correlations between tumor cell percentage and metabolite levels in the CNB specimens were assessed according to the immunohistochemical status of the primary tumor. In univariate analysis, levels of choline-containing compounds, glutamate, glutamine, glycine, serine, and taurine were correlated with primary tumor characteristics. In multivariate analysis, most metabolite levels were not affected by tumor cell percentage. Tumor cell percentage showed poor correlation with metabolite levels in hormone receptor-positive cancer and triple-negative cancer, and poor to fair correlation with metabolite levels in HER2-positive cancer. This study showed that differences in the tumor cell fraction of CNB samples do not affect predictions on the primary cancer from which the samples are obtained.     

Validation of tissue microarrays using p53 immunohistochemical studies of squamous cell carcinoma of the larynx.

Tissue microarrays are a powerful new tissue-conserving technology in the study of cancer, allowing simultaneous study of a large number of tumor specimens. We sought to ascertain the utility of tissue microarrays in head and neck cancer pathology using squamous cell carcinoma of the larynx as a model system. Whole-specimen slides from 44 different laryngeal squamous cell carcinomas were stained for p53 expression. Microarrays were then generated by taking six 0.6-mm core biopsies from each of the 44 specimens. The whole sections and the microarrays were independently scored for p53 expression. Twenty-three (53%) of the 44 tumor specimens were positive for p53. Forty-four of the 264 core biopsies (17%) were not given a score because of the lack of tumor cells. Seventy-eight percent of the individual discs on the microarray had scores in agreement with those of the whole-section slides. Among biopsy discs with tumor cells present, 94.5% were in agreement with the whole-section slide. The average probability that four randomly chosen biopsy discs, considered together, would accurately identify the presence of p53 staining in a whole section was 0.97 (95% CI.93-1.0). We conclude that tissue microarrays for squamous cell carcinomas can accurately represent immunohistochemical results of whole-slide specimens when four or more samples are used. Tissue microarrays are an important technique that may be applied to immunohistochemical studies of head and neck cancer.