Identifying patient segment drug responses – IEEE Life Sciences

By Ram Aiyar and Patrick Meyer, IEEE
IEEE

This is a paradigm of how identifying the “right-drug-for-the-right-patient” would look like in treating a devastating inflammatory disease like Rheumatoid Arthritis.

Rheumatoid Arthritis (RA) is a devastating disease in which the patients have persistent inflammation in their joints. Over time, bone erosion, destruction of cartilage, and complete loss of joint integrity can occur. Eventually, multiple organ systems may be affected. The root-cause of the disease is unknown, but evidence points to a complex interplay between environmental and genetic factors. No single diagnostic test confirms the diagnosis of rheumatoid arthritis. However, several tests can provide objective data that increases diagnostic certainty and allows progression of the disease to be followed. The diagnostics tests include biochemical measurements in blood and urine, white blood and red blood cell counts, and radiographic findings (X-ray) in inflamed joints. Newer methods to provide early diagnosis of RA using Doppler technologies [1] and identification of progression of bone damage in RA [2] are being investigated.

To increase value to the patient as well as third party insurers, pharmaceutical companies are now looking at identifying patient segments that will respond positively to their expensive drugs. Typically, this would involve development of a diagnostic test in combination with development of the drug, a paradigm otherwise known as “Theranostics” [3]. The development of the diagnostic test is usually done as part of the clinical trials process. The schematic shows one approach to the development of the diagnostic test.

Identify the Right Drug

Click to enlarge

B. Biological sample measurements

As mentioned earlier, typically in RA, specific biochemical measurements are made that will help in the diagnosis of the disease. The specific list of biochemical measurements was compiled over years of practicing medicine and sharing information between doctors and researchers. These typically include protein measurements from blood and urine.

With the advent of genetic sequencing techniques, researchers have tools to detect not only hundreds of proteins and other molecules, but also the genetic predisposition of a person from a small biological sample. Currently, we can decode a person’s genetic background using Next Generation sequencing methods [4]. Proteins are synthesized by the activity of specific enzymes on genetic information stored in DNA. A vital intermediary to this process is a family of molecules known as RNA (See video “Transcription and Translation”).

The advent of microarray technology has helped in the measurement of greater than 50,000 RNA molecules from a very small sample. Moreover, mass spectrometry can be used to provide high content protein information from a sample. The above technologies just give a flavor of how much information can be obtained from a biological sample from a patient. These advancements in technologies are particularly useful in RA where there is limited access to human joint tissue from normal and disease subjects [5].

C. Data Storage and Management

This vast amount of valuable data obtained through biological measurements, may have the information to identify a small panel of measurements that will result in a diagnostic tests that could enable identification of patients who will respond to a particular drug. However, managing this vast amount of data and analyzing it is not trivial. Storage, transfer, management and processing of such information are tedious. Researchers in the fields of parallel processing [for example, see 6, 7, 8], nanotechnology [for example, see 9, 10, 11], computer networking [for example, see 12, 13, 14] and cloud computing [for example, see 15, 16, 17] have played a vital role in performing similar functions for data of a different kind, such as audio and video information. Expertise in these fields is essential to manage the data so that analytical methods could be applied in a timely fashion.

D. Computational Methods

Once the data is collected, processed and stored in a manageable fashion, it will be subjected to classification algorithms [for example, see 18] to identify a panel of measurements or markers that is able to separate patients that respond to the drug from those that do not with a greater than 90% accuracy. The fields of pattern recognition [for example, see 19], machine learning [for example, see 20, 21], computer vision [for example, see 22], image processing [for example, see 23, 24] and signal processing [for example, see 25, 26] is more than 100 years old and has much to offer to solve these classification problems. An example of the use of classification algorithms in predicting response to drugs in RA is the use of a random forest predictor [27]. That being said, the inherent complexity of biological systems, the error-prone and incomplete nature of data collection mechanisms, and the large size of biological datasets motivate novel algorithms with significant memory and computational requirements [28]. Because of this, it has been predicted that biology is set to become a highly quantitative science. In the 21st century, biology will become the most computer-intensive science [29].

Hide References
  1. Özkan, A.O., et al. Analysis of Doppler signals of radial artery for diagnosis of rheumatoid arthritis. in Biomedical Engineering Meeting (BIYOMUT), 2010 15th National. 2010. Antalya: IEEE. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5479856&isnumber=5479681.
  2. Leung, K., et al., Automatic quantification of changes in bone in serial MR images of joints. IEEE Trans Med Imaging, 2006. 25(12): p. 1617-26. Available from: http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=4016168.
  3. Vogenberg, F., I.B. C, and M. Pursel, Personalized medicine: part 1: evolution and development into theranostics. Pharmacy and Therapeutics, 2010. 35(10): p. 560-76. Available from: http://www.nxtbook.com/nxtbooks/medimedia/pt_201010/
    index.php?startid=Cover1
    .
  4. Zhao, J. and S. Grant, Advances in Whole Genome Sequencing Technology. Curr Pharm Biotechnol. November 4, 2010. Available from: http://www.ncbi.nlm.nih.gov/pubmed/21050163.
  5. Yeh, C., et al., Current development of saliva/oral fluid-based diagnostics. Tex Dent J., 2010. 127(7): p. 651-61. Available from: http://www.ncbi.nlm.nih.gov/pubmed/20737986.
  6. Müller, T. and A. Knoll. A generic approach to realtime robot control and parallel processing for industrial scenarios. in Industrial Technology (ICIT), 2010 IEEE International Conference on. 2010: IEEE. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5472758&isnumber=5472440.
  7. Takeuchi, Y., et al. Scalable parallel processing for H.264 encoding application to multi/many-core processor. in Intelligent Control and Information Processing (ICICIP), 2010 International Conference on. 2010: IEEE. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5565292&isnumber=5564160.
  8. Yukita, K., et al. Study of AC/DC power supply system with DGs using parallel processing method. in Power Electronics Conference (IPEC), 2010 International. 2010: IEEE. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5544599&isnumber=5542000.
  9. Hodge, A., et al., Catch the Wave-Nanotechnology, the Future is now [In the News]. Engineering in Medicine and Biology Magazine, IEEE, 2010. 29(1): p. 10-15. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5415988&isnumber=5415964.
  10. Guizhi, X., et al., Study on the Feasibility to Detect Cancer Tumors by Combining Nanotechnology With SQUID. Applied Superconductivity, IEEE Transactions on, 2010. 20(3): p. 1956-1959. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5424087&isnumber=5471713.
  11. Weixia, Y., et al., Independent Sets of DNA Oligonucleotides for Nanotechnology Applications. NanoBioscience, IEEE Transactions on, 2010. 9(1): p. 38-43. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5325721&isnumber=5440010.
  12. Fuh-Gwo, C., C. Kuo-Yi, and L. Chin-Yang. A cooperative learning of computer networking with portable laboratories using virtual machine. in Computer Science and Information Technology (ICCSIT), 2010 3rd IEEE International Conference on. 2010. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5564422&isnumber=5563692.
  13. Xiaojuan, L., G. Yong, and W. Minhua. Web-Based Virtual Laboratory Platform for Computer Network. in Computational Intelligence and Software Engineering, 2009. CiSE 2009. International Conference on. 2009. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5365506&isnumber=5362501.
  14. Zhou, Y.-m., C.-y. Xing, and M. Chen. Design and Implement Computer Networking Virtual Laboratory. in Education Technology and Computer Science, 2009. ETCS ’09. First International Workshop on. 2009. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4958977&isnumber=4958969.
  15. Hofmann, P. and D. Woods, Cloud Computing: The Limits of Public Clouds for Business Applications. Internet Computing, IEEE, 2010. 14(6): p. 90-93. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5617066&isnumber=5617049.
  16. Pallis, G., Cloud Computing: The New Frontier of Internet Computing. Internet Computing, IEEE, 2010. 14(5): p. 70-73. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5562494&isnumber=5562482.
  17. Yan, L., Network I/O Virtualization for Cloud Computing. IT Professional, 2010. 12(5): p. 36-41. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5499454&isnumber=5593026.
  18. Benso, A., S. Di Carlo, and G. Politano, A cDNA Microarray Gene Expression Data Classifier for Clinical Diagnostics Based on Graph Theory. Computational Biology and Bioinformatics, IEEE/ACM Transactions on, 2010. PP(99): p. 1-1. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5582075&isnumber=4359833.
  19. Cheng, Q., A Sparse Learning Machine for High-Dimensional Data with Application to Microarray Gene Analysis. Computational Biology and Bioinformatics, IEEE/ACM Transactions on, 2010. 7(4): p. 636-646. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=4770093&isnumber=5614813.
  20. Comaniciu, D. Robust ultrasound image analysis using learning. in Biomedical Imaging: From Nano to Macro, 2010 IEEE International Symposium on. 2010. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5490356&isnumber=5490050.
  21. Dong, Q. and S. Zhou, Novel Non-Linear Knowledge-Based Mean Force Potentials Based on Machine Learning. Computational Biology and Bioinformatics, IEEE/ACM Transactions on, 2010. PP(99): p. 1-1. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5567096&isnumber=4359833.
  22. Huang, C.-H., et al. Bio-inspired computer visual system using GPU and Visual Pattern Assessment Language (ViPAL): Application on breast cancer prognosis. in Neural Networks (IJCNN), The 2010 International Joint Conference on. 2010: IEEE. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5596972&isnumber=5595732.
  23. Kyriacou, E.C., et al., A Review of Noninvasive Ultrasound Image Processing Methods in the Analysis of Carotid Plaque Morphology for the Assessment of Stroke Risk. Information Technology in Biomedicine, IEEE Transactions on, 2010. 14(4): p. 1027-1038. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5444988&isnumber=5505920.
  24. Latifoğlu, F., D. Sönmezer, and K. Deniz. Determination of mitosis cells number using image processing methods. in Biomedical Engineering Meeting (BIYOMUT), 2010 15th National. 2010. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5479823&isnumber=5479681.
  25. Liu, X., et al., Multiple Functional ECG Signal Processing for Wearable Applications of Long-Term Cardiac Monitoring. Biomedical Engineering, IEEE Transactions on, 2010. PP(99): p. 1-1. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5530360&isnumber=4359967.
  26. Yuqi, Z., et al. Study on Noninvasive Auto-Estimation of Pulmonary Vascular Resistance by Doppler Audio-Signal Processing System. in Bioinformatics and Biomedical Engineering (iCBBE), 2010 4th International Conference on. 2010. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5514930&isnumber=5514659.
  27. Bienkowska, J., et al., Convergent Random Forest predictor: methodology for predicting drug response from genome-scale data applied to anti-TNF response. Genomics, 2009. 94(6): p. 423-32. Available from: http://www.ncbi.nlm.nih.gov/pubmed/19699293.
  28. Schmidt, B., S. Aluru, and D.A. Bader, Message from the Workshop Chairs 7th IEEE International Workshop on High Performance Computational Biology (HiCOMB 2008), 2008. Available from: http://www.hicomb.org/papers/HICOMB2008-00.pdf.
  29. Noble, D., The rise of computational biology. Nature Reviews Molecular Cell Biology, 2002. 3(6): p. 495-499. Available from: http://www.ncbi.nlm.nih.gov/pubmed/12042768.