Annals of Surgical Oncology Sign the Guestbook
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/ASO.2005.03.073
Annals of Surgical Oncology 12:943-949 (2005)
© 2005 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stretch, J. R.
Right arrow Articles by Lean, C. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stretch, J. R.
Right arrow Articles by Lean, C. L.

Original Article

Melanoma Metastases in Regional Lymph Nodes Are Accurately Detected by Proton Magnetic Resonance Spectroscopy of Fine-Needle Aspirate Biopsy Samples

Jonathan R. Stretch, DPhil, FRACS1,2, Ray Somorjai, PhD3, Roger Bourne, PhD4, Edward Hsiao, MD4, Richard A. Scolyer, FRCPA5, Brion Dolenko, MSc3, John F. Thompson, MD, FRACS1,2, Carolyn E. Mountford, DPhil4 and Cynthia L. Lean, PhD4

1 Sydney Melanoma Unit and Melanoma and Skin Cancer Research Institute, Royal Prince Alfred Hospital, Camperdown, New South Wales 2050, Australia
2 Department of Surgery, University of Sydney, Camperdown, New South Wales 2006, Australia
3 Institute for Biodiagnostics, National Research Council of Canada, 435 Ellice Avenue, Winnipeg, Manitoba R3B 1Y6, Canada Manitoba Canada
4 Institute for Magnetic Resonance Research and Department of Magnetic Resonance in Medicine, University of Sydney, Block 3, Level 3, Royal North Shore Hospital, St. Leonards, New South Wales 2065, Australia
5 Department of Anatomical Pathology, Royal Prince Alfred Hospital, Missenden Road, Camperdown, New South Wales 2050, Australia

Correspondence: Address correspondence and reprint requests to: Cynthia L. Lean, PhD; E-mail: cynth{at}imrr.usyd.edu.au.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Background: Nonsurgical assessment of sentinel nodes (SNs) would offer advantages over surgical SN excision by reducing morbidity and costs. Proton magnetic resonance spectroscopy (MRS) of fine-needle aspirate biopsy (FNAB) specimens identifies melanoma lymph node metastases. This study was undertaken to determine the accuracy of the MRS method and thereby establish a basis for the future development of a nonsurgical technique for assessing SNs.

Methods: FNAB samples were obtained from 118 biopsy specimens from 77 patients during SN biopsy and regional lymphadenectomy. The specimens were histologically evaluated and correlated with MRS data. Histopathologic analysis established that 56 specimens contained metastatic melanoma and that 62 specimens were benign. A linear discriminant analysis–based classifier was developed for benign tissues and metastases.

Results: The presence of metastatic melanoma in lymph nodes was predicted with a sensitivity of 92.9%, a specificity of 90.3%, and an accuracy of 91.5% in a primary data set. In a second data set that used FNAB samples separate from the original tissue samples, melanoma metastases were predicted with a sensitivity of 87.5%, a specificity of 90.3%, and an accuracy of 89.1%, thus supporting the reproducibility of the method.

Conclusions: Proton MRS of FNAB samples may provide a robust and accurate diagnosis of metastatic disease in the regional lymph nodes of melanoma patients. These data indicate the potential for SN staging of melanoma without surgical biopsy and histopathological evaluation.

Key Words: Melanoma • Lymph node metastases • Magnetic resonance spectroscopy • Cancer • Statistical classification strategy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Staging of patients with melanoma requires accurate evaluation of sentinel nodes (SNs).1 SNs are currently identified by lymphatic mapping and are surgically harvested for detailed histopathologic assessment.2 The SN procedure is invasive and has a 5% to 10% complication rate, including wound infection, lymphocele formation, and, in some cases, allergic reactions to the blue dye. In addition, histopathologic examination of the nodes requires labor-intensive serial sectioning and expensive immunohistochemistry. Accurate nonsurgical SN evaluation would offer significant advantages over SN excision and histopathologic evaluation. This study was undertaken to determine the accuracy of magnetic resonance spectroscopy (MRS) performed on fine-needle aspirate biopsy (FNAB) samples in nodes removed surgically to identify metastatic melanoma and, thereby, establish a basis for the future development of a nonsurgical technique to assess SNs.

Proton MRS reflects the chemical composition of cells and the biochemical changes associated with the disease process.3 MRS can be performed on tissue or FNAB-derived material to determine tissue pathology with sensitivity and specificity rates of approximately 95% when an automated pattern-recognition method is used.3–5 Similarly, MRS studies of urine6 have been successful in predicting pathology and transplanted organ rejection.7

Simple methods of discrimination between spectra characterizing different tissue types have achieved reasonable predictive accuracy for some pathologic processes. For example, measurement of the relative intensities of the citrate and creatine/choline resonances in in vivo spectra of prostate tissue is reported to predict the presence of prostate cancer with approximately 85% accuracy.8 We have previously demonstrated that in melanoma patients, proton MRS of solid tissue or FNAB samples can determine the presence of metastatic melanoma in excised lymph nodes (P < .012) by a simple measurement of the relative intensities of the two broad resonances from lipid/metabolite compounds and choline-containing compounds.9 In this earlier study, only visually obvious peaks were measured in the spectra from a small sample database. A more sophisticated analysis of a larger sample database would be expected to improve predictive accuracy.

For clinical databases, it is usually the case that the number of spectral data points generated by each sample measurement greatly exceeds the number of samples available to be measured. A wide range of automated pattern-recognition methods have been developed for spectroscopy and other information-dense analytical techniques (see reviews10–12). These methods have in common the ability to select from very large data sets the often very subtle features that best discriminate between classes (e.g., pathologic processes) of interest.

In the study reported here, we applied proton MRS and an automated linear discriminant analysis (LDA)-based pattern-recognition method termed the statistical classification strategy (SCS) to a large database of FNAB samples from resected lymph nodes. The robustness of the SCS method has been demonstrated by the analysis of proton magnetic resonance (MR) spectra from several tissues, including thyroid, ovary, prostate, esophagus, liver, breast, and brain tumors.5,13


   PATIENTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 
All histopathologic and MRS analyses were undertaken blind. Correlation of MRS and histopathologic data was made after all reports were filed.

Patients, Specimen Collection, and Specimen Storage
FNAB samples (involving a single cortical puncture followed by multiple passes through each quarter of the specimen by using a 25-gauge needle attached to a 3-mL plastic syringe) were obtained from 118 tissue specimens from 77 melanoma patients undergoing regional node operation for suspected metastatic melanoma (SN biopsy or complete node dissection) in accordance with ethics protocol X97-0103 (Central Sydney Area Health Service, Sydney, Australia). All FNAB samples were placed in polypropylene vials containing 300 µL of phosphate-buffered saline (PBS; .27 mM of KCl, 13.69 mM of NaCl, 1.52 mM of KH2PO4, and 15.2 mM of Na2HPO4; pH 7.2) made up in perdeuterated water (PBS/D2O) and immediately snap-frozen in liquid nitrogen and stored at –70°C (for <6 weeks) until thawed for spectroscopy. Metastatic melanoma (22 microscopic metastases and 34 macroscopic metastases) was identified by histopathologic analysis in 56 of the specimens. These were obtained from 22 SNs, 23 non-SNs, and 11 specimens in which no residual recognizable lymph node component could be identified. Sixty-two specimens were benign by histopathologic analysis. These were obtained from 50 SNs, 11 non-SNs, and 1 biopsy sample in which no residual recognizable lymph node component could be identified.

Magnetic Resonance Spectroscopy
FNAB samples were thawed and transferred directly to 5-mm susceptibility-matched MRS tubes. Proton MRS analyses were performed on a Bruker Avance 360-MHz wide-bore spectrometer (Bruker Biospin BmbH, Rheinstetten, Germany) equipped with a 1H/13C 5-mm probe head. Samples were spun at 20 Hz, and the temperature was maintained at 37°C. One-dimensional spectra were acquired with acquisition parameters as follows: frequency, 360.10 MHz; flip angle, 60° (6- to 7-microsecond pulse length); interpulse delay, 1 second; acquisition time, .8 seconds; 8000 complex pairs of data points; 256 transients; and spectral width, 3600 Hz, with a total acquisition time of 5 minutes. The field was locked to D2O. Water suppression was effected by a selective excitation field gradient method (double-pulse field gradient spin echo14). Chemical shifts of resonances were referenced to water at 4.64 ppm (nominal position with respect to tetramethylsilyl propionate in PBS/D2O at 37°C). Magnitude spectra were created by Fourier transformation of the free induction decay data without application of window functions.

Histopathologic Evaluation
Histopathologic evaluation for the presence of melanoma metastasis was performed on all biopsy samples. Each SN was cut in its longitudinal axis in 3-mm-thick slices and embedded in entirety in paraffin blocks after tissue processing. The histopathology protocol for the detection of metastatic melanoma was based on the protocol for the Multicenter Selective Lymphadenectomy Trial (MSLT) coordinated at the John Wayne Cancer Institute (Santa Monica, CA) and funded by National Cancer Institute Grant PO1 CA29605-12. Four sequential 5-µm-thick tissue sections were cut from each block and stained with hematoxylin and eosin (H&E), S100 protein, or HMB-45. The sections were examined microscopically, initially at a scanning magnification of x100. Positive tissue samples were identified by the presence of metastatic melanoma cells on H&E, by immunohistochemically stained tissue sections, or both. This technique has a reported failure rate of .8%.15 Non-SNs were assessed histopathologically for the presence of metastatic melanoma by assessment of single H&E-stained sections.

Pattern-Recognition Method
Diagnostic correlation was performed between the MRS and histopathology data by using the SCS method,4,5 designed specifically for biomedical spectroscopy databases that contain many fewer spectra than data points in each spectrum. Magnitude spectra consisting of 4096 data points over the spectral width of 10 ppm were reduced to 1500 points between .35 and 4.00 ppm. Outside this region, all resonances have a very low signal-to-noise ratio. The spectra were normalized to the total integral in this region. Both the first derivatives and the rank-ordered first derivatives (i.e., replacing the actual first-derivative values by their positions in the sorted derivative list: the smallest derivative value would receive rank 1, and the largest, rank n, n being the number of derivative values) of the MR magnitude spectra were used for the analysis. Both processed versions of the spectra (first derivative and rank-ordered first derivative) were analyzed by a genetic algorithm-based optimal region selection algorithm (ORS_GA) to identify maximally discriminatory subregions in the spectra. The ORS_-GA used as its objective function the results of a leave-one-out cross-validated LDA classification. Fifteen subregions were chosen by ORS_GA for each of the 2 types of spectral preprocessing (first derivative and rank-ordered first derivative), giving a total of 30 subregions as features for classifier production. From these 30 features, all possible 7-feature subsets were found by exhaustive search, and each was used to produce LDA classifiers. The probabilities produced by the best-balanced eight of these classifiers were used as the input features for the third stage of the SCS: classifier fusion. All LDA classifiers were made robust by bootstrap cross-validation, and the process was repeated 10,000 times. The bootstrapping process minimizes "overtraining" over the developed classifier by randomly selecting half of the sample data, deriving a classifier from these data, and testing the accuracy of this classifier on the remaining half of the sample data. The final LDA classifier was based on the accuracy-weighted average of the 10,000 different bootstrap classifier coefficient sets. Samples are classified according to class-assignment probability. Classification is termed crisp if the assignment probability is ≥75% and fuzzy if it is <75%. Note that development of the classifier is based on repeated validation with data independent of the training data. The overall performance of the final classifier is assessed by classification of the entire primary data set and a secondary validation set.

The primary data set comprised MR spectra of FNAB samples from 56 tissue samples containing metastatic melanoma and 62 benign samples. A secondary validation set comprising duplicate FNAB samples from a subset of the same tissue samples used for the primary data set (24 melanoma containing and 38 benign) were also classified with the classifier developed from the primary data set. This secondary validation set tests the variability of FNAB samples from the same tissue, variability in sample handling and storage, and variability within the proton MRS measurement procedure.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Typical proton MR spectra of FNAB samples from histologically benign and malignant specimens are shown in Figure 1Go. Resonances in the spectra include those consistent with the methyl and methylene (CH2) protons of lipid at .9 and 1.3 ppm, respectively, and the N-trimethyl of choline and choline-based metabolites at 3.2 ppm. Additional resonances include those consistent with acyl chain protons (-CH=CH-CH2-) and N- or O-acetyl groups at 2.0 ppm; creatine, phosphocreatine, and lysine at 3.0 ppm; taurine at 3.4 ppm; inositol at 3.6 ppm; and protons of carbohydrate residues in the 3.5- to 4.0-ppm region.


Figure 1
View larger version (11K):
[in this window]
[in a new window]
  		FIG. 1. One-dimensional proton magnetic resonance spectra (8.5 T) of fine-needle aspirate biopsy samples of regional lymph nodes from melanoma patients: (A) node containing metastatic melanoma and (B) benign node. CH3, methyl; CH2, methylene; +N(CH3)3, N-trimethyl of choline and choline-based metabolites; -CH=CH-CH2- acyl chain protons, N- or O-acetyl groups at 2.0 ppm; Cr, creatine; PCr, phosphocreatine; Lys, lysine; Tau, taurine; inos, inositol; CHOH, carbohydrate residues.

 
Pattern-recognition analysis was applied to the two histopathologic categories, i.e., benign tissues and tissues containing metastatic melanoma. In the primary data set, the presence of metastatic melanoma was predicted with a sensitivity of 92.9%, a specificity of 90.3%, and an accuracy of 91.5%. Six false positives and four false negatives were identified. In the secondary validation set (using second FNAB samples from a subset of the same tissues), the presence of metastatic melanoma was predicted with a sensitivity of 87.5%, a specificity of 90.3%, and an accuracy of 89.1%. The crispness of the data was calculated as the percentage of samples assigned to a class with a class probability ≥75%. Samples that could not be assigned with this probability were termed fuzzy. The crispness of the data was 93.2% for the primary data set (8 of 118 samples were fuzzy) and was 92.7% for the secondary validation set (4 of 62 samples were fuzzy). These results are listed in Table 1Go.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Statistical strategy–based classification of magnetic resonance spectroscopy data of fine-needle aspirate biopsy samples from regional lymph nodes of melanoma patients
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
CONCLUSIONS
 REFERENCES
 
This study establishes the capacity of MRS performed ex vivo on FNAB samples from surgically harvested SNs to accurately diagnose metastatic melanoma. It is known that MRS monitors biochemical changes, including those related to cellular proliferation16 and cell-surface fucosylation,17,18 that occur throughout tumor development and progression. It is the identification of these biochemical changes that forms the basis of this technique. MRS is of proven accuracy in its ability to diagnose cancers in many organs3,5,16,19–31 and to identify breast cancer metastases in lymph nodes in humans19 and an animal model.16 In the case of breast and thyroid cancers, the MRS diagnosis of malignancy has been made, as in this study, on FNAB samples of the tested tissue.19,25

The combination of FNAB and MRS illustrates the potential use of new techniques to determine a patient’s SN status. It is reasonable to consider the future development of minimally invasive assessment of SNs by using ultrasound-directed needle aspiration. One potential strategy might consist of conventional lymphoscintigraphy followed by percutaneous ultrasound-guided FNAB of the identified SNs. Each FNAB could be initially assessed by MRS for spectral features associated with the presence of metastatic tumor. Ultrasound guidance would provide the primary validation of the correct localization of the SN. If mapping agents containing antimony were used, assay of antimony levels could be used to further validate the biopsy results of the appropriate node. We have recently established that ion coupled plasma-mass spectroscopy assay of antimony, a rare metal derived from one of the available lymphoscintigraphy agents, can be used to confirm the accuracy of SN identification.32,33

A rapid, accurate, cost-effective, nonsurgical method of evaluating nodes for their tumor-harboring status would greatly improve the treatment of melanoma patients by obviating the need for invasive SN excision and histopathologic assessment. This would not only reduce patient morbidity, but also produce significant cost savings.

The use of an automated pattern-recognition analysis of MRS data provides a more accurate method for the diagnosis of metastatic melanoma than was possible by the relatively simple comparison of resonance-intensity ratios.9 This was also the case for MRS-based diagnosis of breast cancer.19

In this study, four false-negative results were identified. On review of the histopathologic data, three of the four false-negative nodes were found to harbor only small amounts of metastatic disease that comprised <2% of the nodal volume. In addition, two of the nodes were of "horseshoe" morphology, containing large fatty hila and slender peripheral lymphoid components. This morphology may have resulted in suboptimal sampling of the nodal tissue, thus causing a false-negative result. It is also appropriate to note that horseshoe nodes present a similar problem when assessed by conventional histologic analysis. The slender peripheral rim of nodal tissue is inevitably not comprehensively examined, because it is impractical to examine any SNs by serial sectioning of the entire node.

Six false-positive SNs were also identified in this study. Because the protocol used for histopathologic assessment involves the microscopic examination of only a small portion of the lymph node tissue (<1%), these cases could have contained metastatic disease that was not detected by histopathologic analysis. Alternatively, MRS may have identified metastatic disease not discernible by histopathologic analysis. In a previous study16 using a rat model for breast cancer metastasis, MRS detected lymph node metastases before they could be identified by histopathologic analysis. The presence of these metastases was confirmed by xenografting the nodal tissue into nude mice and documenting tumor growth.

The SN status of a melanoma patient has been established as the single most important determinant of survival, ranking above the Breslow thickness of the primary tumor, the presence of ulceration, and the mitotic rate.34 Although a definitive answer to the question of whether subsequent early completion regional node resection benefits patients with positive SNs is still awaited with the results of the Multicenter Selective Lymphadenectomy Trial, knowledge of the SN status is critical to the accurate stratification of patient risk cohorts in adjuvant therapy trials. This study demonstrates that MRS combined with an automated pattern-recognition analysis can reliably identify metastatic melanoma in the regional lymph nodes by analysis of FNAB samples.

MRS on lymph nodes from melanoma patients reported in this series was performed on FNAB samples obtained from surgically harvested tissues. However, MRS on FNAB samples obtained during conventional investigation of clinically palpable nodes is also under way at our institution. It may be feasible to develop clinical techniques that combine conventional lymphoscintigraphy and MRS of ultrasound-guided percutaneous FNAB samples of SNs to avoid surgical node biopsy and its potential complications.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
REFERENCES
 
Proton MRS of FNAB samples from regional lymph nodes obtained surgically provides for accurate diagnosis of metastatic disease in melanoma patients. Proton MRS, therefore, has the potential to determine, rapidly and objectively, the lymph node status of patients with melanoma. This study demonstrated high levels of sensitivity and specificity for detecting metastatic melanoma in lymph nodes. With the development of ultrasound-guided percutaneous FNAB sampling of SNs, this technology may be adapted to provide accurate but minimally invasive assessment of regional lymph node status.


   ACKNOWLEDGMENTS
 
Supported by Australian National Health and Medical Research Council Grant 107226. The authors thank Chitra De Silva and Lawrence Li from the Sydney Melanoma Unit for collection of samples and pathology data. They also thank those at the Institute for Magnetic Resonance Research: Susan Dowd for storing and archiving the specimens and Sinead Doran for invaluable help in preparing this manuscript.

Received for publication April 19, 2004. Accepted for publication June 29, 2005.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 CONCLUSIONS
 REFERENCES
 

  1. Thompson JF, Uren RF. Lymphatic mapping and sentinel node biopsy for melanoma. Exp Rev Anticancer Ther 2001;1:446–52.
  2. Thompson JF, McCarthy WH, Bosch CM, et al. Sentinel lymph node status as an indicator of the presence of metastatic melanoma in regional lymph nodes. Melanoma Res 1995;5:255–60.[Medline]
  3. Mountford CE, Doran S, Lean CL, Russell P. Cancer pathology in the year 2000. Biophys Chem 1997;68:127–35.[Medline]
  4. Somorjai R, Dolenko B, Nikulin A, et al. Distinguishing normal allografts from biopsy-proven rejections: application of a three-stage classification strategy to urine MR and IR spectra. Vibr Spectrosc 2002;28:97–102.[CrossRef]
  5. Lean CL, Somorjai RL, Smith ICP, Russell P, Mountford CE. Accurate diagnosis and prognosis of human cancers by proton MRS and a three-stage classification strategy. In: Webb GA, ed. Annual Reports on NMR Spectroscopy, Volume 48. Guildford, UK: Academic Press, 2002: 71–111.
  6. ‘t Hart B, Vogels J, Spijksma G, Brok H, Polman C, van der Greef J. 1H-NMR spectroscopy combined with pattern recognition analysis reveals characteristic chemical patterns in urines of MS patients and non-human primates with MS-like disease. J Neurol Sci 2003;212:21–30.[Medline]
  7. Rush D, Somorjai R, Deslauriers R, Shaw A, Jeffery J, Nickerson P. Subclinical rejection—a potential surrogate marker for chronic rejection—may be diagnosed by protocol biopsy or urine spectroscopy. Ann Transplant 2000;5:44–9.[Medline]
  8. Swanson MG, Vigneron DB, Tran TK, Kurhanewicz J. Magnetic resonance imaging and spectroscopic imaging of prostate cancer. Cancer Invest 2001;19:510–23.[Medline]
  9. Lean C, Bourne R, Thompson J, et al. Rapid detection of metastatic melanoma in lymph nodes using proton magnetic resonance spectroscopy of fine needle aspiration biopsy specimens. Melanoma Res 2003;13:259.[CrossRef][Medline]
  10. Reo NV. NMR-based metabonomics. Drug Chem Toxicol 2002;25:375–82.[CrossRef][Medline]
  11. Hagberg G. From magnetic resonance spectroscopy to classification of tumours. A review of pattern recognition methods. NMR Biomed 1998;11:148–56.[CrossRef][Medline]
  12. El-Deredy WA. Pattern recognition approaches in biomedical and clinical magnetic resonance spectroscopy. NMR Biomed 1997;10:99–124.[CrossRef][Medline]
  13. Doran S, Falk G, Somorjai R, et al. Pathology of Barrett’s oesophagus by proton magnetic resonance spectroscopy and a statistical classification strategy. Am J Surg 2003;185:232–8.[Medline]
  14. Braun S, Kalinowski HO, Berger S. 150 and More Basic NMR Experiments: A Practical Course. 2nd ed. Weinheim, Germany: Wiley-VCH, 1998.
  15. Li XL, Scolyer RA, Ka VSK, et al. Pathological review of negative sentinel lymph nodes in patients with regional recurrence of melanoma: a clinicopathologic review of 1170 cutaneous melanoma patients undergoing sentinel lymph node biopsies, including recommendations for the pathological examination of sentinel lymph nodes. Am J Surg Pathol 2003;27:1197–202.[Medline]
  16. Mountford C, Lean C, Hancock R, et al. Magnetic resonance spectroscopy detects cancer in draining lymph nodes. Invasion Metastasis 1993;13:57–71.[Medline]
  17. Lean CL, Mackinnon WB, Delikatny EJ, Whitehead RH, Mountford CE. Cell-surface fucosylation and magnetic resonance spectroscopy characterization of human malignant colorectal cells. Biochemistry 1992;31:11095–105.[CrossRef][Medline]
  18. Mackinnon WB, Huschtscha L, Dent K, Hancock R, Paraskeva C, Mountford CE. Correlation of cellular differentiation in human colorectal carcinoma and adenoma cell lines with metabolite profiles determined by 1H magnetic resonance spectroscopy. Int J Cancer 1994;59:248–61.[Medline]
  19. Mountford CE, Somorjai RL, Malycha P, et al. Diagnosis and prognosis of breast cancer by magnetic resonance spectroscopy of fine-needle aspirates analysed using a statistical classification strategy. Br J Surg 2001;88:1234–40.[CrossRef][Medline]
  20. Delikatny EJ, Russell P, Hunter JC, et al. Proton MR and human cervical neoplasia 1: ex vivo spectroscopy allows distinction of invasive carcinoma of the cervix from carcinoma in situ and other preinvasive lesions. Radiology 1993;188:791–6.[Abstract/Free Full Text]
  21. Mackinnon WB, May GL, Mountford CE. Esterified cholesterol and triglyceride are present in plasma membranes of Chinese hamster ovary cells. Eur J Biochem 1992;205:827–39.[Medline]
  22. Kuesel AC, Kroft T, Saunders JK, Prefontaine M, Mikhael N, Smith IC. A simple procedure for obtaining high-quality NMR spectra of semiquantitative value from small tissue specimens: cervical biopsies. Magn Reson Med 1992;27:349–55.[Medline]
  23. Russell P, Lean CL, Delbridge L, May GL, Dowd S, Mountford CE. Proton magnetic resonance, human thyroid neoplasia. I: Discrimination between benign and malignant neoplasms. Am J Med 1994;96:383–8.[CrossRef][Medline]
  24. Delbridge L, Lean C, Russell P, et al. Proton magnetic resonance and human thyroid neoplasms: potential avoidance of surgery for benign follicular neoplasms. World J Surg 1994;18:512–7.[Medline]
  25. Lean CL, Delbridge L, Russell P, et al. Diagnosis of follicular thyroid lesions by proton magnetic resonance on fine needle biopsy. J Clin Endocrinol Metab 1995;80:1306–11.[Abstract]
  26. Lean CL, Newland RC, Ende DA, Bokey EL, Smith IC, Mountford CE. Assessment of human colorectal biopsies by 1H MRS. correlation with histopathology. Magn Reson Med 1993;30:525–33.[Medline]
  27. Mackinnon WB, Russell P, May GL, Mountford CE. Characterisation of human ovarian epithelial tumours (ex vivo) by proton magnetic resonance spectroscopy. Int J Gynaecol Cancer 1995;5:211–21.[Medline]
  28. Hahn P, Smith ICP, Leboldus L, Littman C, Somorjai RL, Bezabeh T. The classification of benign and malignant human prostate tissue by multivariate analysis of H-1 magnetic resonance spectra. Cancer Res 1997;57:3398–401.[Abstract/Free Full Text]
  29. Mackinnon WB, Barry PA, Malycha PL, et al. Fine-needle biopsy specimens of benign breast lesions distinguished from invasive cancer ex vivo with proton MR spectroscopy. Radiology 1997;204:661–6.[Abstract/Free Full Text]
  30. Bezabeh T, Smith IC, Krupnik E, et al. Diagnostic potential for cancer via ‘H magnetic resonance spectroscopy of colon tissue. Anticancer Res 1996;16:1553–8.[Medline]
  31. Bezabeh T, Samy E, Patel R, et al. Statistical classification strategy for proton magnetic resonance spectra of soft tissue sarcoma: an exploratory study with potential clinical utility. Sarcoma 2002;6:97–103.[Medline]
  32. Dawson M, Doble P, Beavis A, Li LX, Soper R, Scolyer RA. Antimony by ICP-MS as a marker for sentinel lymph nodes in melanoma patients. Analyst 2003;128:217–9.[CrossRef][Medline]
  33. Scolyer RA, Thompson JF, Li LXL, et al. Antimony concentrations in nodal tissue can confirm sentinel node identity. Mod Pathol 2004;17:1191–7.[CrossRef][Medline]
  34. Balch CM, Soong SJ, Gershenwald JE, Thompson JF, Reintgen DS, Cascinelli N. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001;19:3622–34.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Ann. Surg. Oncol.Home page
A. Beavis, M. Dawson, P. Doble, R. A. Scolyer, R. Bourne, L.-X. L. Li, R. Murali, J. R. Stretch, C. L. Lean, R. F. Uren, et al.
Confirmation of Sentinel Lymph Node Identity by Analysis of Fine-Needle Biopsy Samples Using Inductively Coupled Plasma-Mass Spectrometry
Ann. Surg. Oncol., March 1, 2008; 15(3): 934 - 940.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Stretch, J. R.
Right arrow Articles by Lean, C. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Stretch, J. R.
Right arrow Articles by Lean, C. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS