Probe into Error in Khoo Teck Puat Hospital's Laboratory Test for Cancer Patients

Ms Joan Pereira asked the Minister for Health with regard to the case of breast cancer patients at Khoo Teck Puat Hospital who may have received unnecessary treatment since 2012 due to inaccurate test results (a) what are the key revelations from the review of the process; (b) what measures have been implemented to prevent similar incidents; and (c) whether there have been cases of patients suffering from detrimental effects.

Dr Tan Wu Meng asked the Minister for Health with regard to cancers inaccurately classified HER2-positive by Khoo Teck Puat Hospital laboratory (a) how many patients (i) received unnecessary treatment and (ii) suffered significant or long-term side effects thereby; (b) what is the ensuing cost to affected patients; (c) what is being done to help these patients; and (d) what measures are taken to ensure fidelity of laboratory tests.

Miss Cheryl Chan Wei Ling asked the Minister for Health with regard to the incident at Khoo Teck Puat Hospital where breast cancer patients received unnecessary treatment since 2012 (a) why did subsequent follow-up checks on the patients over the years not pick up any misdiagnosis from the earlier test; and (b) whether there are regular audits of hospitals and medical clinics on laboratory protocols to identify lapses.

Ms Hazel Poa asked the Minister for Health what are the reasons for the wrong test results for cancer patients at Khoo Teck Puat Hospital and what measures will be put in place to ensure that such errors do not occur again.

Sir, on 19 November 2020, Khoo Teck Puat Hospital (KTPH) was informed by its laboratory that its immunohistochemistry (IHC) tests for Human Epidermal Growth Factor Receptor 2 (HER2) were producing higher-than-expected rates of positive results for breast cancer patients. Preliminary investigations by the laboratory suggested that some of the HER2 results may be inaccurate.

Following the incident, MOH has been working closely with KTPH to ensure that affected patients are provided with adequate support. KTPH has sent the samples of all patients, who have been tested HER2 positive since 2012, when HER2 testing first started in KTPH, to external laboratories for re-testing to determine how many have received inaccurate results. Preliminary investigations by the KTPH laboratory suggest that the inaccurate results could be due to a suboptimal staining process. The KTPH laboratory has since stopped in-house testing of HER2. MOH has also issued an alert to our other public healthcare institutions to conduct a quick review of their laboratory-developed IHC tests to ensure that positivity rates are within the acceptable range. Thus far, we have not received any reports of similar risks from other healthcare institutions.

KTPH's Department of Laboratory Medicine is subject to regular inspection by MOH as part of regulatory processes under the Private Hospitals and Medical Clinics (PHMC) Act to ensure that its laboratory facilities, systems and processes are in place to meet patient and personnel safety standards. In addition, the laboratory is accredited by the College of American Pathologists (CAP), where the last biennial inspections by peers were conducted in 2019.

As at 23 December 2020, 200 patients have been reclassified from HER2 positive to HER2 negative. Of these, eight patients were treated at private hospitals and 192 patients at Government hospitals. Eight patients are still pending retests. Joint care teams have been formed, comprising KTPH surgeons, histopathologists and the treating oncologists, to review the individual care plans for these affected patients, based on the change in their HER2 status.

KTPH and the treating oncologists are in the process of actively reaching out to these patients to conduct open disclosure and assess these patients for any potential side effects due to unnecessary treatment. The more common side effects include diarrhoea, chills and fatigue – these are usually short-lasting. About 3% to 4% of those who underwent HER2-directed treatment, for example, using Herceptin, may also experience heart problems. KTPH is also reviewing the bills of these affected patients. The portion of the bills which arose from the unnecessary treatment will be fully refunded. KTPH is also ready to provide any clinical and financial support to the affected patients including on-going or follow-on treatments, if any, which may be needed as a result of this over-treatment.

The National Healthcare Group has convened an independent review committee, comprising external experts from multiple relevant disciplines in the healthcare industry. The objective is to conduct a thorough evaluation of the incident, to understand better the lapses that have occurred and recommend appropriate measures to improve the process. This ensures that any system gaps are identified and addressed swiftly to prevent recurrence of similar incidents.

The committee's investigations are on-going right now and more time would be required to ensure a thorough review. NHG will provide an update when more information is available and these findings will be shared with the other healthcare institutions for improvement.

Mr Speaker, I start by declaring that I am a doctor who happens to be a specialist in medical oncology but I should also declare that my Clementi residents too follow the news. Some have also raised this issue with me, in my Member of Parliament capacity.

Sir, a number of residents have said this unfortunate situation should not have happened. But it has, and it is incumbent upon the authorities and the hospital to do the very best to support the affected patients in as holistic a way as possible and as compassionately as possible. Sir, I have two supplementary questions for the Senior Minister of State.

The first is that in terms of supporting the patients, will the relevant authorities bear in mind that patients who undergo unnecessary treatment as a result of this unfortunate situation, some of these patients will have had to make recurrent visits to clinic, recurrent visits for treatment involving time-off work, care-givers taking time-off work – all of these with possible financial implications as well as implications for job security in the current COVID-19 climate. And I ask, will the Ministry, together with KTPH, consider this compassionately and holistically in supporting the patients who have gone through this?

The second question to the Senior Minister of State is that even as the root cause analysis is on-going, can the Ministry assure us that we will also look at whether any areas of improvement may impinge on other immunohistochemistry tests that may have come out of that relevant laboratory?

Sir, I say this because pathology tests, immunohistochemistry tests – they are like the charts which guide sailors at sea, which guide the treating medical team, which guide the patient. And if there has been an issue with the charts for whatever reason, it is, of course, a concern for navigating the patient care journey and giving the relevant treatment. Sir, I raise these questions to the Senior Minister of State.

Mr Speaker, I thank the Member, Dr Tan, for raising these two questions that I am sure, must be on the minds of many patients and relatives who are cancer sufferers.

First, I must say that Herceptin treatment for HER2 positive patients usually is not so much of a monotherapy, that means on a standalone basis. Usually, it is in combination with other chemotherapy drugs for breast cancer. So, in other words, even in a HER2 negative patient, where you do not need Herceptin treatment, you will still require some chemotherapy to overcome breast cancer, and therefore, you will still have to come for recurrent visits, treatment, monitoring and testing by the managing oncologist. But we know that some of these can create a lot more anxiety, maybe a few extra visits that will be required for monitoring of side effects.

KTPH and the managing oncologists and doctors will take a compassionate approach to reassure, to also help these patients to overcome their disease and anxieties by keeping close communication with them through a more dedicated care team, especially those who are affected by the change in result status.

But I must also say that in this situation where the issue is more of a false positive, it does mean that some patients would have perhaps, over treatment. I think it is probably a preferred state than to have under treatment, especially when you have cancer. In this situation, HER2 testing has shown that some of them should have been classified as HER2 negative, meaning that actually the patient does have a better prognosis compared to those who are HER2 positive.

Overall, it is not necessarily a bad thing for the patients to be actually reclassified as HER2 negative on testing. KTPH has informed that any costs, as well as investigations that are done as a result of Herceptin treatment will be borne by the hospital and refunds will be given. So, I think that will go in some way to relieve some of the financial pressures and stresses that a patient has undergone.

On the second question of whether other immunohistochemistry tests would be reviewed; I think we must understand that immunohistochemistry testing is classified in most laboratories as a highly complex test. By highly complex, what I mean is that it is a test that is not quite the same as you drawing a blood test – pop it into the machine and the result comes up within a few minutes. It is not as simple as that. It is a multi-step process that has multiple steps of human intervention in the process.

First, as a tumour specimen is removed, what we call cold ischemic time – how soon the tumour, after it is removed from the body – gets put into a fixation, that means to fix the tissue in the chemical, usually it is recommended to be less than one hour. So, how the operating team in the Operating Theatre, not even the pathologist, just how the operating team handles the tissue – when do they take a specimen out, when do they call for the team to put into the solution; you must also put into the correct solution, otherwise if you put in the wrong chemical to fix the tumour tissue, that sample cannot be used for IHC staining.

Then, when it goes to the laboratory, how thick the specimen is sliced, what kind of stains are used, what kind of temperature that is done, what is the concentration of the stains being used, what is the choice of antibody and subsequently after staining, how the pathologists themselves grade the degree of staining. And the degree of grading or staining is what we call a semi quantitative method. It means it is not a "plus" or "minus", "yes" or "no". It is a gradation of degrees of staining just like when you look at your iPhone photos, and you try to adjust the contrast. Between two sets of persons looking at the adjustment, to you, this may look pleasing enough, dark enough, but to another person, this may be too light.

So, there is some degree of inter-observer variation which trained pathologists must then make the judgement call and say is this dark enough stain to be deemed as a positive 3+, 2+, 1+ or 0. So, it is semi-quantitative in that sense.

Therefore, you can appreciate that this being a very complex test, as in most other immunohistochemistry tests, will require multiple levels of validation. But I will also say that the reason why this issue was discovered is because the institutional processes within KTPH have allowed this to be picked up. Because being a multi-step process, if we depend on one single point to validate its accuracy, then it will be at a very, very high risk of failure at some point of time. But fortunately, because there is an institutional process through the Tumour Board – a multi-disciplinary review board looking at tumour outcomes, specimen and treatment, it was flagged up at the point of review that the positivity rate seems to be higher than expected. And that was what triggered the review.

I would say that the processes are there. We now need to look at which part of this process could have been strengthened further and that is the on-going review that the expertise team is looking at, at the moment. We will also use this chance to learn about some of the failures that may have happened in this multi-step process and help the laboratory to strengthen them for other immunohistochemistry tests as well.

Thank you, Mr Speaker. I would like to ask the Senior Minister of State because my original questions were not being answered.

Actually, the main thing around it is this, because as the Senior Minister of State has said, all these patients had gone for recurrent visits to their respective doctors. Certainly, over the entire period of time, something must have been picked up even if the first diagnosis was supposed to be wrong. So, as the Senior Minister of State said, with all these specimens and also all the different processes that are in place, why was there not an earlier trigger? I think that is the key to the issue.

Sir, I thank the Member for asking this question. It is going to be a little bit technical but let me attempt to explain this in a way that most people can understand.

Inherent in any laboratory test, there is always a chance of false positivity, false negativity. That relates to the sensitivity and specificity of the test. In recent months, because of COVID-19, we have all been doing PCR testing, everybody is a little bit more attuned to the concept of why a certain test can give you a false positive and false negative result.

Immunohistochemistry test actually is fairly notorious for having this kind of risk because, as I said, it is semi-quantitative, so the staining process is like mixing poster colours. If any part of the mixing process – the process of processing the specimen by putting stains, dyes and counter-stains – is not done optimally, the key word here is "optimal", you can perform all the steps but if the timing is a bit off, maybe the concentration is off by a little bit, the kind of picture that comes out on the slide that the pathologist gets to look at, will be quite different in terms of contrast and stain.

If I may quote from a paper in Journal of Clinical Oncology published in 2015, a British medical journal, it gives a background that early studies using this technique to look at HER2 staining has shown high false positivity rate, as high as 19% – as high as 19% – in the early days when this test was started. Over the years, after 2007, when a stronger push by the professionals to put out guidelines, that false positivity rate has now dropped to about 6% or less, but it is still present; it is never going to be zero.

What we need to do, as in this instance, where the clinical practice within an institution does regular review of data accumulated over time; because to show that there is now an over-representation over time, you need to have accumulated enough caseload over a period of time, to show on a retrospective basis that after three years, five years, I accumulated 200 patients, 1,000 patients, and I aggregate the data and I realise, "Oops, it is actually more than 6%". Then, this is when it triggers a review.

But when you first start the test, you only have one case, two cases, three cases, there is not enough to do statistical analysis to tell you whether you are above or below the 6% mark.

So, I hope that gives a sense that it is not a complete failure of the system; it is inherent in the test and that is how a review process through a multi-disciplinary tumour board picks up any outlier conclusions to trigger a review.

I thank the Senior Minister of State for his reply. I have a few supplementary questions.

Firstly, while these errors were actually discovered in a review process, since the error actually started in 2012, why did it take so long to discover the errors? Is it a standard timeframe for enough statistics to flag up a possible error? Secondly, would the Ministry consider a second independent test for serious illnesses? The last supplementary question is about compensation to the victims. What sort of compensation is being considered and is a lifetime of free healthcare on the cards?

Sir, let me clarify the Member's misconception. When I said in my reply that KTPH is reviewing all the results from 2012, 2012 was taken at a point in which the test first started in KTPH. It does not mean that all the results going back to 2012 are erroneous. So, in that sense, KTPH is being prudent and careful, super kiasu, going back all the way to when the test first started. It may well be that the errors could have occurred in the last three months or six months.

I think the key point usually is when the reagent has been changed. A different brand of reagent or a different provider of the reagent could usually be the starting point of why the protocol or the test conditions were no longer as optimal as they should be.

But going back to 2012 when the test first started was really trying to be comprehensive and to be careful, to make sure that we do not miss anybody in the process. It does not mean that the test was erroneous from 2012. The results of the review are still on-going, so I do not want to prejudge the review process from the experts. Let us wait for the experts to look at all the results going back to 2012 and then decide at which point, what step of the process that could have been where the error occurred.

When it comes to the question of whether you are going to get compensation, as I said earlier, any treatment cost due to the treatment arising from a positive HER2 results, in other words, usually treatment with Herceptin as the drug, the cost of the drug likely would be refunded. Any investigation or tests or treatment of any related complications as a result of, say, Herceptin, would also be refunded as well. But to then say that this translates to a lifetime of compensation, I think that would be a little bit too far a thing to stretch because most of the complications, if any, or most of the side effects relating to the treatment, are transient – diarrhoea, a bit of fever, a bit of chills. They do not have long-term consequences, they are not long-lasting and we have to take that in the correct context of medical treatment.

I hope that answers the Member's questions.

I also had a question about a second independent test for serious illnesses.

I take that as a general question which in any form of medical diagnostics, if the clinician is unsure that a particular test alone cannot give you a high degree of probability of predicting a positive diagnosis or presence of an illness, then it behoves clinical judgement to call for a second test. But immunohistochemistry is a specific unique test. It does not necessarily mean that an alternative is available. Although in the literature, there are other alternative tests that could be done but we have to consider that against the time needed to do it because that could delay treatment, and also whether that brings on unnecessary costs.

That, I think, we should leave it to the experts to come up with the process guidelines on what is a testing algorithm to decide for Herceptin-related treatment.

Thank you, Mr Speaker. I think the Senior Minister of State also have not fully answered my Parliamentary Question regarding, how many patients were actually involved and how many patients have MOH engaged in the first place currently now that are affected by this particular error. Currently, what measures have been taken to ensure the well-being of the patients, both psychologically as well as physical health? I would also like to know what was the average cost of the medication, of this particular medicine that was just mentioned. Because the Senior Minister of State did not mention anything about the costs; he just mentioned that it will be refunded but I am not sure what is the actual cost.

I do have a patient, who is my resident in Sembawang GRC, who saw me at my Meet-the-People Session. She was very worried when she saw the news; she said she is one of them. When I asked her whether has MOH contacted her, she said not yet. So, she was asking me what legal help can she now seek. So, I would like to ask the Senior Minister of State what avenues do affected patients have to seek compensation for such an error?

Sir, the anxieties of patients are quite understandable, especially when they read about this in the news.

I would want to give assurance and say that if you have not been contacted by your oncologist, it means you are okay. Those who are affected, as I said in my reply, 200 patients so far have been reclassified from HER2 positive to HER2 negative. In other words, these 200 patients from a poor prognosis tumour now has a better prognosis tumour, because HER2 positivity codes for a poorer prognosis to begin with.

Therefore, it needs the managing oncologists to go on a one-to-one with any of these patients whose status have been reclassified, to do a detailed conversation on their own clinical prognosis, what is the stage of the tumour, what treatment has already been given in the past, what will be the on-going treatment needed.

Not all patients, I must say, who have been tested HER2 positive be it now or previously, will go on to Herceptin treatment because it is a treatment decision that is guided by the needs of the patient, not just by the test alone. There are some patients for which the body may be too frail; even if you are HER2 positive, you may not receive Herceptin treatment. On the other hand, there are recurrent tumours where the oncologist feels that you have reached end of line, you will still require Herceptin treatment as a last resort, even though it may not be the first line in some other cases.

The decision matrix is not a straightforward one. For patients who may be affected, it is best to seek your oncologist's advice. But if you have not been contacted specifically for this, it means that you are okay, your status remains as it was from the last conversation with your oncologist.

The cost will vary from institution to institution. I do not have the exact numbers right now because it also depends on how many doses, how long the duration of treatment each patient got, and also different body size of patients may require different dosages to begin with. So, that is something that we cannot generalise; it is probably quite individual. That is why the dedicated care team from KTPH, especially the oncologists, will have to reach out to affected patients one by one and go through the conversation in a detailed manner.

Thank you, Speaker. I thank the Senior Minister of State for his reply. In addition to the physical side effects that many of these patients might have suffered, some of them might suffer panic attacks, anxiety attacks. So, would Khoo Teck Puat Hospital be able to support medical costs relating to mental health?

Sir, I think it is best we leave such a assessment and decision to the managing joint care team together with the patient. A panic attack or anxiety episode can occur with or without cancer, can occur with or without treatment. There can be a myriad of factors involved. But the joint care team will deal with the patients with compassion and find a way to help them overcome this current hurdle they are facing.