Slide 1

Thank you very much Chairman good afternoon everyone. That is the question.

Slide 2

But I want to answer two questions, firstly is tonsillectomy a logical treatment? And secondly is it an effective treatment?

Slide 3

Let me remind you that the IgA in your circulation is mostly monomer but there is some polymer in which 2 monomers are joined by the bridging protein J chain. Whilst in your mucosal fluid you have secretory IgA in which polymers of IgA bind the secretory component and that IgA as you’ve already heard comes in two subclasses IgA 1 and IgA2.

Slide 4

The mucosal immune system produces polymeric IgA which reaches mucosal fluids as secretory IgA and in health little or none of that polymeric IgA reaches the circulation. Your circulating IgA is bone marrow derived, mostly monomeric, mostly IgA 1.

Slide 5

We’ve known for a long time that the IgA that’s in the kidney is polymeric and IgA1.

Slide 6

There are very few studies to suggest secretory IgA is ever in the kidney.

Slide 7

As has already been discussed today people began by believing that it was likely that the abnormality in this disease was in the mucosal immune system. Let’s understand the mucosal immune system a little better and the role the tonsil plays.

Slide 8

The main effector site in your mucosa where polymeric IgA is produced and then reaches mucosal fluids is the mucosal lamina propria and as Rosanna Coppo told you this is an enormous area, the whole of your mucosa is very large indeed. That is the main effector site.

Slide 9

Here is an example of that this is --- mucosa each of these red cells is a polymeric IgA producing cell and you can imagine the enormous area you have in your mucosa producing polymeric IgA.

Slide 10

But also in the mucosal immune system are inductive sites the gut, bronchial and mucosal associated lymphoid tissue and these are organised lymphoepithelial tissues.

Slide 11

They do produce themselves small amounts of polymeric IgA which can reach mucosal fluids or potentially reach the circulation.

Slide 12

But the main role of these tissues is to populate the lamina propria with immunocompetent primed B cells.

Slide 13

You’ll be familiar with Peyer’s patches as an example of gut associated lymphoid tissue.

Slide 14

But for our purposes today we’re interested in mucosal associated lymphoid tissue which includes the tonsils and the adenoids and a number of other structures, for example, the ring of lymphoepithelial tissue around the larynx known as Waldeyer’s pharyngeal ring.

Slide 15

What happens if you remove the tonsils? You have not removed most of this organised lymphoepithelial tissue. You will prevent the production of any IgA of a specific type being produced by the tonsil but you will not prevent this process.

Slide 16

Sure enough the effect of a tonsillectomy is to remove one source of polymeric IgA producing cells sure but clinical mucosal immunity is not compromised and that seems pretty likely because there’s clearly enormous redundancy in this system, both bronchial and gut and in the upper respiratory tract as well and other lymphoepithelial tissues remain.

Slide 17

What do we know about the tonsils which have been removed from patients with IgA nephropathy? Well, there are some data which suggest that those tonsils which have been removed are unusually producing more polymeric IgA than normal tonsils and that’s based on evidence of an increased number of IgA producing cells in those tonsils that this is polymeric IgA producing cells, antigen presenting cells within the tonsil are unusually focused on IgA 1 with an increase of IgA 1 over IgA2. So there is some evidence that the tonsils that are being removed may favour polymeric IgA production.

Slide 18

But I think you have to understand that for tonsillectomy to be an effective treatment in IgA nephropathy it will only be effective if that tonsil is exclusively producing the pathogenic IgA and the rest of your mucosal immune system is not.

Slide 19

First question of a clinical sort is it possible to get IgA nephropathy after a tonsillectomy? The answer, of course, is yes and I suspect that all of you have seen such a case.

Slide 20

The other issue we have to consider is that although discussion began that mucosa was the source of the IgA that reaches the kidney in this condition it’s already been mentioned to you today by Rosanna Coppo that there is much data to suggest we should look elsewhere.

Slide 21

There are studies suggesting that in fact polymeric IgA production is reduced in the lamina propria in the IgA nephropathy in the mucosa and increased in the bone marrow both on the basis of cell counting and also on the basis of immunization studies.

Slide 22

Furthermore there is evidence that if you have a mucosal antigen in IgA nephropathy you do not have the normal situation so consistently whereby a systemic immune response has diminished or prevented the phenomenon of oral tolerance because normally mucosal antigens are dealt with there.

Slide 23

But in fact there is some evidence that when mucosal antigens are encountered in IgA nephropathy it is a systemic marrow derived response which is increased.

Slide 24

Now of course we have no proof that IgA nephropathy is a single disease. Several of these mechanisms might work to result in IgA deposition in the kidney. Remember also that we have no proof that IgA nephropathy is the same disease in all parts of the world and that is important when we come to look at the clinical literature about tonsillectomy.

Slide 25

So why did tonsillectomy become a treatment for IgA nephropathy? Well, clearly some people believe that mucosal immunity was important.

Slide 26

It was also very clear that there were episodes of visible hematuria which seem to be provoked by upper respiratory infection and in some patients by recurrent tonsillitis and therefore, that you could reduce those episodes by a tonsillectomy.

Slide 27

But of course, that doesn’t prove the tonsils are the source of the pathogenic IgA necessarily and I don’t think we yet fully understand the relationship between an upper respiratory tract infection for example, and the episode of hematuria that follows it.

Slide 28

There’s another very interesting but rare clinical observation which you ought to think about as well. I’ve seen this once, here’s a patient having a tonsillectomy and this is the urine of that patient one hour after they got back to the surgical ward. I’ve seen it once and I know that others have seen it as well. I don’t know what that means but if you squeeze the tonsil, in some people they get hematuria. It’s not much to do with an IgA mediated immune response and I don’t know what the mechanism is and I encourage you to just think about that because clinical observation often challenges us with difficult questions.

Slide 29

So, that’s the background. Make of that what you will as to whether you believe a tonsillectomy is a logical thing to do. Nevertheless people have done it and it’s important that we now assess the published evidence.

Slide 30

How would we measure success?

Slide 31

Well, you might measure it by reducing episodes of macroscopic hematuria and of course, you may be able to do that and we can consider that.

Slide 32

You might assess it by the improvement of urinary abnormality, reductions in microscopic hematuria for example, or proteinuria.

Slide 33

Best of all you might want to know if you could prevent end stage renal disease.

Slide 34

But a fourth question also might be as to whether you can prevent transplant recurrence. If a patient had a tonsillectomy but still got renal failure. Would you prevent the recurrence by having a tonsillectomy having had a tonsillectomy?

Slide 35

You need to put that in the context of the risks of this procedure. This is UK tonsillectomy audit data which reminds us that when a child has a tonsillectomy they have a 1% chance of a complication in hospital including blood transfusions and a return to surgery and a 4% chance of readmission to hospital many of those with secondary haemorrhage. And that this is an operation with a  mortality of between 1 in 15.000 and 1 in 25.000.

Slide 36

Available data also suggests that if you do tonsillectomies in adults, the post tonsillectomy haemorrhage rate is twice that in adults compared to children. So we have to bare that in mind when we assess this as a possible treatment.

Slide 37

We have no randomised controlled trials of tonsillectomy in IgA nephropathy.

Slide 38

We are reliant on retrospective observational studies that creates us some problems.

Slide 39

First of all controls are very variable in those studies. In some of them they are historic controls, in others they are contemporaneous controls and they vary in size and number.

Slide 40

Secondly criteria for tonsillectomy are varied and not always easy to be clear about.

Slide 41

In most of those studies the tonsils have been taken out because they were obviously infected and the patient had chronic recurrent tonsillitis. But in some of those studies clinically normal tonsils were being taken out because it was thought to be therapeutic. There has been a fashion for some sort of tonsillar provocation test whereby the tonsils would be stimulated and evidence of changes in urine findings, for example, would be used to indicate whether tonsillectomy would be likely to help. But I think it’s fair to say that those tests are mostly now discredited.

Slide 42

Because these are retrospective observational studies the use of other treatments in these studies is also very variable. So the use of corticosteroids, of cyclophoshamide, of anticoagulants, of antiplatelet agents is very variable in both the patients themselves and in control patients, as is the use of blockade of the renin angiotensin system as is the blood pressure target to which these patients were controlled.

Slide 43

So then let’s see what we can do about measuring success. Yes these data do tell us that tonsillectomy will reduce episodes of macroscopic haematuria not to 0 a typical figure is that 60% reduction will occur but nevertheless you can reduce those episodes. We have already had some thought about how important that is.

Slide 44

What about tonsillectomy improving urinary abnormalities? What data do we have on that?

Slide 45

Well here are the data. There are seven studies which address this issue. You’ll notice that 6 of them come from Japan, studies mostly from the 1990s. Here are the number of patients undergoing a tonsillectomy in each of those studies, you’ll see two at the bottom really very large studies and that these studies have a mean follow up which is relatively short between one and 6 years. They describe urinary remission which is usually defined as the absence of proteinuria and the disappearance of microscopic haematuria or urinary improvement which is variously defined.

Slide 46

One of those studies simply shows the reduction in proteinuria in the whole group but the rest of the studies show that sort of expectation of remission of urinary abnormality between 30 and 60%. Most of those studies if they make further analysis say that remission is more likely at what is described as early disease. That is to say patients with less than 1 g of proteinuria a day typically or those in whom histological injury is described as mild. Think about that for a moment because we all know that spontaneous remission does occur in IgA nephropathy. I’m sure you’ve all seen it.

Slide 47

Between 20-30 % of our patients with microscopic haematuria only that disappears with follow up. Such spontaneous remissions which I’ve been observing in my clinic for a long time are much more likely to occur with what you might call early disease.
Of course nowadays if we use a blockade of the renin-angiotensin system very actively, we can often induce remission that is to say the disappearance of proteinuria with very low grade proteinuria if we use those treatments.

Slide 48

So what about the second question? Does tonsillectomy prevent end stage renal disease?

Slide 49

We have 4 studies that address that question two from Europe and this is one of them from Germany, a small study in which 16 patients had a tonsillectomy and they were compared to 39 who did not and they were followed over more than 10 years. This is the probability of renal survival and you will see that there is no difference in renal survival between the two groups.

Slide 50

This is a study from Japan, a study in which 48 patients had a tonsillectomy and 70 did not. This is again renal survival and you’ll see the curve showing a separation appearing with those who had a tonsillectomy beginning to show superior outcome to those who did not. It’s more than a decade before that difference happens.

Slide 51

This is a good example of a study where large numbers of patients in this study both receiving tonsillectomy or not had very many different additional medications. In that study they go on to look at hazard ratios using a Cox regression model to identify which factors if any were associated with a significant reduction in a hazard ratio of less than one. They describe in this study that only tonsillectomy has such a significant effect more so than steroid therapy, the presence of interstitial lesions on histology or a mean arterial pressure of less than 95.

Slide 52

Here is the most recent study from Nanjing in China published just this year comparing 54 who had a tonsillectomy and 58 who did not. You may begin to see a separation between these lines in your own mind but it is not statistically significant even at 15 years.

Slide 53

Let me put together for you the four studies which address this question. The study from Hungary which I’ve not shown you data from yet, from Germany, from Japan and from China. These are the numbers of individuals who have undergone tonsillectomy. These are the mean follow up periods, at least 10 years in all cases. In only one of those, the study that goes right through to a mean follow up of 15 years is the statistical evidence of the prevention of end stage renal disease but I remind you again that these are all retrospective and observational studies.

Slide 54

Finally, how to measure success what about this question about whether we will prevent transplant recurrence with a tonsillectomy and we have absolutely no data on that rather interesting question.

Slide 55

I remind you again that we have no proof that IgA nephropathy is a single disease and it is perfectly possible that there are different mechanisms resulting in IgA deposition and no proof that in the end when we understand this disease better we will really regard it as a single disease in all parts of the world.

Slide 56

So there could be a specific group with tonsillar abnormalities who benefit from tonsillectomy. What is certain is that we do not yet know how to identify such a group.

Slide 57

A randomised controlled trial would of course be highly informative. I’m given to understand that there is such a trial underway in Japan. I do not know the protocol of that trial but clearly the results of that would be extremely interesting and might change practice.

Slide 58

But for the moment let’s return to the question is tonsillectomy an effective therapeutic option for IgA nephropathy?

Slide 59

I’ve shown you the available evidence.

Slide 60

I’m pretty clear. No. I’m absolutely clear about the answer. Thank you for your attention.