Discriminant analysis of trace element distribution in normal and malignant human tissues

Area of Science:

• Biochemistry and Molecular Biology
• Medical Diagnostics
• Analytical Chemistry

Background:

• Accurate classification of malignant versus normal human tissues is crucial for diagnosis and treatment.
• Trace element profiles in tissues may differ significantly between healthy and cancerous states.
• Ultramicro energy dispersive X-ray fluorescence (UMEDXRF) offers a sensitive method for elemental analysis.

Purpose of the Study:

• To evaluate the potential of trace element analysis via UMEDXRF for classifying human breast, colon, and lung tissues.
• To determine the accuracy of discriminant analysis in differentiating malignant from normal tissues.
• To identify specific trace elements most effective in tissue-type and malignancy classification.

Main Methods:

• Measured levels of 16 trace elements in malignant and histologically normal human breast, colon, and lung tissues.
• Employed ultramicro energy dispersive X-ray fluorescence (UMEDXRF) for precise elemental quantification.
• Utilized discriminant analysis to create linear composites of trace elements for classification.

Main Results:

• High classification accuracies were achieved: 98% for breast, 100% for colon, and 100% for lung tissues.
• Optimal malignant-normal discrimination involved different sets of trace elements for each tissue type (e.g., Ca, Rb, Zn for breast).
• Three-group tissue discrimination (breast, colon, lung) reached 85% accuracy with paired tissues and 91% with tumor tissues alone.

Conclusions:

• Discriminant analysis of trace element levels is a highly effective methodology for classifying human tissues.
• UMEDXRF provides a valuable tool for distinguishing malignant from normal tissues across different organs.
• Specific trace element signatures can reliably differentiate between breast, colon, and lung cancers.