timfong888/text.txt

## text.txt
The Problem
The number of donor organs available far exceeds the number of patients requiring a kidney (7800 patients waiting for an organ with just 2695 transplants taking place in the UK 2010/11). Therefore it is essential to utilise donor organs wisely. The best match between donor and recipient depends upon genetic similarity (tissue-type) and antibody status (which antibodies have been found to other peoples tissue-types)
Patients may have antibodies which will preclude them from even being considered for a kidney.
The problem we have at present is that not all antibodies defined seem to correlate with poor outcome. In other words, the tests we use to identify antibodies are far too sensitive but we have to be cautious and include them on a patient’s record. Furthermore antibody detection can vary for the same sample depending on which laboratory tests them.
Thus by comparing actual transplant successes and failures with the data produced by the tests, we are able to better correlate which antibody and at what level, is clinically significant. This information will then allow us to adjust our arbitrary ‘cut-off’ values accordingly to define a level at which we can call an antibody significant enough to consider.
I have recently had two Abstract published in Transplant International Volume 24, Supplement 2 , Sep 11. One was in collaboration with 3 other labs about how the antibody detection test results can differ in antibody strength between individual laboratories, even though the same samples were tested by each. (MFI Values Don’t Travel p.357).
Another compared antibody detection tests from 2 manufacturers, again showing differences in which antibodies were found and their relative strength (SAB Normal and Clean Bead Comparison p.353).
This latter abstract was also accepted by the American Society of Histocompatibility and Immunogenetics.

So why hasn’t this been done?
In short, it is being done. All laboratories already compare transplant outcomes to antibody test data to set cut-off points. However the information from the tests are usually held in disparate systems and therefore require laborious efforts to pull together the information. Furthermore, the test that dictates whether a transplant can go ahead (Crossmatch test) usually has a cut-off level that has been defined on data from years ago. Therefore few laboratories continually assess the relevance of these cut-off values because of the large amount of effort required each time. It is known that low level antibodies are not a contraindication to transplantation. Therefore as treatments have got better, the cut-off values for the crossmatch test may be misleading. In other words a crossmatch result just above the cut-off point to define positivity would mean a transplant will not go ahead. However the result may now be considered to be low risk (or even a false positive result) as the cut-off value was originally based upon old data that has not been recently evaluated against recent antibody level and transplant outcome data.

The Solution
The idea is basically for a software system that can compare values from 3 parameters (tests). Two sets of data will be continuous numerical data (an antibody detection test, and a crossmatch test) and one binary (Transplant Success/Failure).
The binary result is the one which ultimately dictates whether a kidney transplant has survived or not. The binary data is simply compared to the numerical data and correlations obtained to investigate cut-off points in the arbitrary continuous scales of the crossmatch and antibody detection tests.
As the antibody detection test is not perfect for quantitative measurements it will be useful to monitor the assays overall performance.
Furthermore, as more research is done to improve the antibody detection test, the improved resolution should be able to be incorporated into the algorithm and eventually allow clearer correlations.

The Opportunity
To date no single software package is used by Transplant Laboratories to compare transplant outcomes to the crossmatch and antibody detection tests. Therefore all current software systems are either home-grown which simply store information, or indeed are paper-based whereby patient details  such as transplant status and even tissue-type are all stored in filing cabinets.
One system which would not only store patient data (eg tissue-type, antibodies that are present, transplant outcome, etc) but be able to incorporate functionality to interrogate the data and produce correlation statistics between the parameters would revolutionise the way Transplants Laboratories work. It would not only provide a powerful audit tool but would also be a powerful research tool.
	The Problem
	The number of donor organs available far exceeds the number of patients requiring a kidney (7800 patients waiting for an organ with just 2695 transplants taking place in the UK 2010/11). Therefore it is essential to utilise donor organs wisely. The best match between donor and recipient depends upon genetic similarity (tissue-type) and antibody status (which antibodies have been found to other peoples tissue-types)
	Patients may have antibodies which will preclude them from even being considered for a kidney.
	The problem we have at present is that not all antibodies defined seem to correlate with poor outcome. In other words, the tests we use to identify antibodies are far too sensitive but we have to be cautious and include them on a patient’s record. Furthermore antibody detection can vary for the same sample depending on which laboratory tests them.
	Thus by comparing actual transplant successes and failures with the data produced by the tests, we are able to better correlate which antibody and at what level, is clinically significant. This information will then allow us to adjust our arbitrary ‘cut-off’ values accordingly to define a level at which we can call an antibody significant enough to consider.
	I have recently had two Abstract published in Transplant International Volume 24, Supplement 2 , Sep 11. One was in collaboration with 3 other labs about how the antibody detection test results can differ in antibody strength between individual laboratories, even though the same samples were tested by each. (MFI Values Don’t Travel p.357).
	Another compared antibody detection tests from 2 manufacturers, again showing differences in which antibodies were found and their relative strength (SAB Normal and Clean Bead Comparison p.353).
	This latter abstract was also accepted by the American Society of Histocompatibility and Immunogenetics.

	So why hasn’t this been done?
	In short, it is being done. All laboratories already compare transplant outcomes to antibody test data to set cut-off points. However the information from the tests are usually held in disparate systems and therefore require laborious efforts to pull together the information. Furthermore, the test that dictates whether a transplant can go ahead (Crossmatch test) usually has a cut-off level that has been defined on data from years ago. Therefore few laboratories continually assess the relevance of these cut-off values because of the large amount of effort required each time. It is known that low level antibodies are not a contraindication to transplantation. Therefore as treatments have got better, the cut-off values for the crossmatch test may be misleading. In other words a crossmatch result just above the cut-off point to define positivity would mean a transplant will not go ahead. However the result may now be considered to be low risk (or even a false positive result) as the cut-off value was originally based upon old data that has not been recently evaluated against recent antibody level and transplant outcome data.

	The Solution
	The idea is basically for a software system that can compare values from 3 parameters (tests). Two sets of data will be continuous numerical data (an antibody detection test, and a crossmatch test) and one binary (Transplant Success/Failure).
	The binary result is the one which ultimately dictates whether a kidney transplant has survived or not. The binary data is simply compared to the numerical data and correlations obtained to investigate cut-off points in the arbitrary continuous scales of the crossmatch and antibody detection tests.
	As the antibody detection test is not perfect for quantitative measurements it will be useful to monitor the assays overall performance.
	Furthermore, as more research is done to improve the antibody detection test, the improved resolution should be able to be incorporated into the algorithm and eventually allow clearer correlations.

	The Opportunity
	To date no single software package is used by Transplant Laboratories to compare transplant outcomes to the crossmatch and antibody detection tests. Therefore all current software systems are either home-grown which simply store information, or indeed are paper-based whereby patient details such as transplant status and even tissue-type are all stored in filing cabinets.
	One system which would not only store patient data (eg tissue-type, antibodies that are present, transplant outcome, etc) but be able to incorporate functionality to interrogate the data and produce correlation statistics between the parameters would revolutionise the way Transplants Laboratories work. It would not only provide a powerful audit tool but would also be a powerful research tool.