Bath Cancer Research Drug Sensitivity Testing Frequently Asked Questions (FAQs)
Comments and further questions welcome (email)
Does Drug Sensitivity Testing work? *
How long does TRAC take? When will I get results? *
How much does TRAC cost? *
What if the test doesn’t work? *
My patient has an ‘odd’ diagnosis. Can you still test their cells? *
My patient has a mixed diagnosis. Can you still test their cells? *
I am having a lymph node excised for diagnosis. I am not sure if the patient has a lymphoma but, if so, this may be the only opportunity to get malignant cells. What should I do? *
I have stored cells from this patient. Can you test them? *
I live abroad, can you still test my cells? *
Can you test solid tumours? *
What does DSI mean? *
Does a DSI of 95% mean that the patient is almost certainly going to respond? *
Why do you give me a result with all the other patients you have tested? I only want one result. *
Why do you show some drugs as sensitive when I know the patient is clinically resistant to them? *
From this TRAC result, how do I work out whether my patient will respond to a drug or not? *
How can I work out whether a patient will respond to a combination of drugs when you only test single agents? *
What about synergism in combinations? *
What is the DiSC assay? Why have you changed to TRAC? *
What about drugs requiring activation in the patient? For instance cyclophosphamide? *
Why do you not test drugs such as rituximab and thalidomide? *
Can you test new or experimental drugs? *
Can you test for side-effects of drugs? *
What alternatives are there to doing a drug sensitivity test? *
Does Drug Sensitivity Testing work?
Yes. Over thirty years of sustained research has shown the same thing, that, in general, if you compare patients treated with test-sensitive drugs with those given test-resistant drugs, the former respond better and survive longer. In 1994, a review of this data [Bosanquet 1994] showed that about 70% of patients treated with test-sensitive drugs responded; whereas less than 10% of patients responded when treated with test-resistant drugs. Where measured, survival was also significantly longer for patients treated with test-sensitive drugs.
How long does TRAC take? When will I get results?
It takes about a week. After we receive the specimen of cells, we work on it for a couple of hours, incubate for 4 days and then score the results and calculate the report (a further 1-2 days). Test results are usually posted, with a brief emailed summary.
How much does TRAC cost?
From £95.00 to £490.00 depending on the number of drugs tested and whether it is private or NHS. Some tests are free, including those done for local patients.
What if the test doesn’t work?
Very occasionally (~2%) the test does not produce results. There may be too few tumour cells in the specimen or they have died during incubation. Also occasionally there are problems in transportation to our laboratory: for example, more than two days transit, too hot or too cold. Care should be taken to ensure that we get a good quality sample in a timely manner. There is no charge for a test that produces no results – we will ask for a further sample.
My patient has an ‘odd’ diagnosis. Can you still test their cells?
Yes, as long as you can tell us what the malignant cells look like. We believe that TRAC is particularly useful in these cases as it is often less clear how the patient should best be treated.
My patient has a mixed diagnosis. Can you still test their cells?
Yes, as long as you can tell us what the malignant cells look like. If we can distinguish one population of malignant cells from the other, we can, if helpful, produce a separate report for each cell type.
I am having a lymph node excised for diagnosis. I am not sure if the patient has a lymphoma but, if so, this may be the only opportunity to get malignant cells. What should I do?
Send us part of the lymph node anyway. We will perform TRAC and wait to hear from you which, if any, are the malignant cells.
I have stored cells from this patient. Can you test them?
Generally no. We need live cells so we cannot test fixed or frozen tissues. Very occasionally, live cells have been stored frozen in special solutions from which they can be retrieved live. We have some success with testing these; please contact us first.
I live abroad, can you still test my cells?
Yes. We need them couriered within 24 hours (48 hours maximum). We cannot test specimens from the USA, but there are a number of commercial laboratories available that do (see list at bottom of the page http://www.weisenthal.org/)
Can you test solid tumours?
We are able, but we do not have sufficient experience to interpret the results. We concentrate on testing leukaemias and lymphomas and this includes testing cells from any site that contains malignant cells: blood, lymph nodes, ascites, pleural fluid or tumour masses from these diseases.
What does DSI mean?
Drug Sensitivity Index. For each drug, the TRAC result is measured in micrograms per millilitre (ug/ml). These figures are meaningless to all but the most experienced: a value of 1ug/ml only has meaning if you know it in the context of other similarly performed tests. Two examples: almost all tests with cladribine give a result considerable less than 1ug/ml, so this value of 1ug/ml indicates significant resistance to cladribine for this patient – a DSI of 5% (only 5% of patients were more resistant). However, most tests with methylprednisolone give a result of over 1ug/ml, so this value of 1ug/ml indicates quite sensitive – a DSI of 85% (only 15% of patients were more sensitive). Thus, a DSI puts this particular patient’s drug result into the context of all the other results with the same drug. We have defined the DSI so that 0% indicates very resistant to the drug and of 100%, very sensitive. Note: this is a laboratory test result, not a probability that the patient will respond.
Does a DSI of 95% mean that the patient is almost certainly going to respond?
A DSI of 95% means that this patient’s cells are sensitive to the drug in the laboratory. It does not tell us a probability of response. In general, if a result shows 'test-sensitive' to a standard drug, we have found approximately 70% of patients will respond. Putting it another way, a patient given this drug is 7 times more likely to respond than if they were resistant to it. There are two things here. Firstly, there are a number of reasons why the patient might not respond to a drug despite a good test result (occasionally the drug might be inactivated or excreted quickly, it may not get to the relevant cells very efficiently, it may not be a very good drug against this disease etc.). Secondly, how accurate a test is at predicting response is also dependent on the probability of the patient responding if you did not do the test. This is loosely referred to as Bayes’ theorem [see Figure 5 on page 12 of Fruehauf & Bosanquet, 1993]. From this graph, you can see that if we have tested a drug which is known to have a response rate of 50%, then a sensitive test result raises the probability of response to about 80%.
Why do you give me a result with all the other patients you have tested? I only want one result.
For each drug, we do only give the one result for your patient. The DSI is a score for this particular patient and this particular drug put into the context of all other tests. (See What does DSI mean? above.)
Why do you show some drugs as sensitive when I know the patient is clinically resistant to them?
There are a number of reasons for a drug being active in the laboratory whilst at the same time not being active in the patient. See above (Does a DSI of 95% …).
From this TRAC result, how do I work out whether my patient will respond to a drug or not?
No amount of laboratory testing will tell you whether a patient will definitely respond or not. On average with TRAC or DiSC, patients given a drug which has been shown to be sensitive in our laboratory test are about 7 times more likely to respond than if they were given a ‘resistant’ drug [see Bosanquet 1994].
How can I work out whether a patient will respond to a combination of drugs when you only test single agents?
It would be best to choose a combination of drugs that includes as many ‘sensitive’ drugs as possible.
What about synergism in combinations?
Some drug combinations are known to be synergistic but probably very few. There would be too many possible combinations to test all in the laboratory, so we stick to testing single agents. It is then the responsibility of the doctor to come up with a good combination based on their knowledge, results from TRAC and consideration of potential drug synergies.
What is the DiSC assay? Why have you changed to TRAC?
The two are almost identical, but TRAC is easier to perform. The results are the same. DiSC stands for Differential Staining Cytotoxicity assay. However, we no longer differentially stain, so we needed a new name. TRAC stands for Tumour Response to Anti-neoplastic Compounds.
What about drugs requiring activation in the patient? For instance cyclophosphamide?
We use activated forms of the drugs. In place of cyclophosphamide, we use mafosfamide in the laboratory. Cyclophosphamide is changed in the body by the liver to the same molecules to which mafosfamide changes in the test-tube.
Why do you not test drugs such as rituximab and thalidomide?
Because they are not cytotoxic in the patient – they use other methods of killing cells (maybe including direct cytotoxicity) and TRAC currently only measures cytotoxicity. Drugs that kill by ADCC (antibody-dependent cell-mediated cytotoxicity) or compliment fixation, for example, may not ‘work’ in a cytotoxicity (cell death) test in the laboratory. Drugs that work in the body by inhibiting the growth of blood vessels will obviously not ‘work’ in a cytotoxicity test (but see Larry Weisenthal’s development of a microvascular viability assay [Weisenthal 2007]). At sufficient concentrations, these drugs may kill cells in the laboratory but the results are unlikely to be relevant to what would happen in the patient.
Can you test new or experimental drugs?
Yes. We are always testing a number. However, new legislation in the UK has made it more difficult to report the results.
As an aside, new drug screening could usefully be performed in the laboratory to identify which malignancies are most likely to benefit from a new drug [Bosanquet et al. 1997].
One of our early tests identified an end-stage AML patient sensitive to fludarabine. At the time, the drug had to be obtained from the NCI in the USA. The patient achieved a partial response [Parker et al. 1992].
Can you test for side-effects of drugs?
No. We have done some experiments looking at the difference in drug sensitivity between tumour and normal cells [Bosanquet, 2004] but we have not determined whether the drug sensitivity of a patient’s normal cells would predict for any side effects of the drug.
What alternatives are there to doing a drug sensitivity test?
There are no direct alternatives to the results obtained from these tests. They produce information that cannot be gained from doing any other laboratory test – the results are independent prognostic markers, giving added value.
For further useful FAQs about drug sensitivity testing, see http://www.weisenthal.org/faqw.htm and