A rapid diagnosis of cancerous tumors is key to improving patient care outcomes.1 Histopathology – which often involves performing a surgical biopsy and then analyzing the tissue via microscopy – is often considered the ‘gold standard’ of cancer diagnosis.2
Image Credit: Giovanni Cancemi/Shutterstock.com
However, there are several drawbacks to using histopathology as a diagnostic technique. The biopsy process is highly invasive and, while the risks are small, it is possible for the patient to develop an infection or bleed from the associated surgical process. Biopsies can also be slow to process as they are dependent on hospital laboratory facilities and personnel availability and their use in diagnosis is dependent on the sampling quality.3
Despite the quality of information that can be obtained from biopsy samples, such an invasive method is unsuitable for tumor monitoring during treatment as it limits the sampling frequency that can be performed. Instead, there has been a great deal of activity developing non-invasive diagnostic and monitoring approaches that can offer a similar quality of information without the patient discomfort and relatively long processing times.
Spectroscopic methods, including Raman spectroscopy, are ideal for the non-invasive imaging of tumors. Spatially resolved versions of the techniques can be used to profile the full tumor site and the spectral information provided at each imaging point can then be decomposed to provide chemical information on the local region.
Challenges of Spectroscopic Techniques for Tumor Monitoring
The human body is an immensely complex environment from a chemical perspective. There are many different types of tissue with different densities and chemical compositions as well as a large amount of water. Skin and other tissues absorb a large amount of incident light, particularly for wavelengths that are in the shorter range of the visible spectrum. This leads to the so-called ‘biological window’, spanning from 650 to 1350 nm, that covers which spectral wavelengths have good penetration depth in tissue and do not undergo undue amounts of scattering.
As most tumors are quite deep-seated, one of the challenges for spectroscopic techniques to be used for tumor monitoring is finding a way for the light to reach the tumor site.
One approach is to use endoscopic methods where fiber optic cables can deliver and detect light from the tumor.5 Recent developments in Raman spectroscopy have made the sampling times sufficiently short that real-time diagnosis can be made using such methodologies and there is hope that these point-of-care methods will be used during live surgical procedures to provide continual information to aid surgeons. This includes live diagnostics for whether all the cancerous cells in the tumor site have been removed.
Other approaches include trying to develop methods that exploit the biological window for spectroscopic sensing, including finding ways to use wavelengths that extend into the near-infrared. These have the greatest amount of tissue penetration and the reduction in scattering also enhances the contrast between the desired image and the background material.
Image Recognition
Spectroscopic diagnostic and imaging methods used in situ do not necessarily offer huge advantages over histopathology if the information obtained remains as complex and time-consuming to interpret. One way to circumvent this problem is to use image recognition methods that can automate the diagnostics process.
Advances in machine learning-based approaches have made it possible to almost fully automate tumor diagnosis with Raman spectroscopy.6 The sensitivity of Raman spectroscopy to the chemical structure of species such as nucleic acids, proteins, lipids, and carbohydrates means that this information can be compared against known disease markers and a diagnosis can be made in a rapid, automated fashion.
The sensitivity of Raman to a variety of chemical species also offers an exciting prospect for more complex tumor immunology. In recent work, engineers have developed non-invasive Raman probes to try and capture a more complete picture of what happens in a tumor environment, particularly in response to immunological treatments.7
Understanding Immunotherapy
Tumors have their own complex microenvironment which does not consist of just the cancerous cells but also blood vessels, immune cells, and other inflammatory cells. Immunotherapy uses mimics of own our immune system and processes to try and fight cancerous cells. In many circumstances, our own immune systems can destroy cancerous cells before tumor masses start to grow.
While new therapies such as monoclonal antibodies seem very promising and may offer a less destructive alternative to radiotherapy and chemotherapy-based treatments, one of the challenges for clinicians is that the dosing has to be unique to each patient.
The teams’ work on developing live Raman probes with automated machine learning for image processing means that it is possible to look at the different interactions between the immunotherapy treatment and the tumor site. From this, it is possible to see how immunotherapy is disrupting the local microenvironment of the tumor and whether the treatment is effective. This information can then be used to adapt treatment courses as necessary.
The chemical identification power of Raman spectroscopy makes it a powerful tool for both diagnostics and understanding how various therapies work and, when combined with automated approaches, it is likely to play a key role in administering and monitoring personalized medicines for cancer treatments.
References and Further Reading
- Neal, R. D. (2009). Do diagnostic delays in cancer matter? British Journal of Cancer, 101, 9-1S2. https://doi.org/10.1038/sj.bjc.6605384
- Irshad, H., Member, S., Veillard, A., Roux, L., & Racoceanu, D. (2014). Methods for Nuclei Detection , Segmentation , and Classification in Digital Histopathology : A Review — Current Status and Future Potential. IEEE Reviews in Biomedical Engineering, 7, 97–114. https://doi.org/10.1109/RBME.2013.2295804
- Pritzker, K. P. H., & Nieminen, H. J. (2019). Needle Biopsy Adequacy in the Era of Precision Medicine and Value-Based Health Care. Arch Pathol Lab Med, 143, 1399–1415. https://doi.org/10.5858/arpa.2018-0463-RA
- Pahlow, S., Weber, K., Popp, J., Bayden, R. W., Kochan, K., Rüther, A., ... & Bhargava, R. (2018). Application of vibrational spectroscopy and imaging to point-of-care medicine: A review. Applied spectroscopy, 72(101), 52-84.
- Duraipandian, S., Bergholt, M. S., Zheng, W., Ho, K. Y., Teh, M., Yeoh, K. G., So, J. B. Y., Shabbir, A., & Huang, Z. (2021). Real-time Raman spectroscopy for in vivo , online gastric cancer diagnosis during clinical endoscopic examination. Journal of Biomedical Optics, 17(8), 081418. https://doi.org/10.1117/1.jbo.17.8.081418
- Hollon, T. C., Pandian, B., Adapa, A. R., Urias, E., Save, A. V, Khalsa, S. S. S., Eichberg, D. G., Amico, R. S. D., Farooq, Z. U., Lewis, S., Petridis, P. D., Marie, T., Shah, A. H., Garton, H. J. L., Maher, C. O., Heth, J. A., Mckean, E. L., Sullivan, S. E., Hervey-jumper, S. L., … Orringer, D. A. (2020). Near real-time intraoperative brain tumor diagnosis using stimulated Raman histology and deep neural networks. Nature Medicine, 26(January). https://doi.org/10.1038/s41591-019-0715-9
- Paidi, S. K., Troncoso, J. R., Raj, P., Diaz, P. M., Ivers, J. D., Lee, D. E., ... & Barman, I. (2021). Raman spectroscopy and machine learning reveals early tumor microenvironmental changes induced by immunotherapy. Cancer Research, 81(22), 5745-5755. https://cancerres.aacrjournals.org/content/81/22/5745
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.