A joint Congress by ERA-EDTA and ISN

LONG-TERM KIDNEY GRAFT SURVIVAL - ARE WE DOING ANY BETTER?

Gerhard Opelz, Heidelberg, Germany

Chair: Michael J. Mihatsch, Basel, Switzerland

Mai Ots, Tartu, Estonia

Prof Gerhard Opelz Department of Transplantation Immunology University of Heidelber Heidelberg, Germany

Slide 1

Doctor Ots, Doctor Mihatsch thank you for your kind introduction. Ladies and Gentlemen I would like to thank the organisers for inviting me. I suppose the reason they asked me to address this topic is they reacted to the same statements in the literature and meetings that I reacted during the last 5 years.

Slide 2

Something like this and this just came across my desk last week by chance. Doctor Matas from Minnesota wrote this in an overview article that he wrote for a Brazilian journal of transplantation as an introduction. I’ve seen his introduction very similar statements in many, many publications. What he says is that in the last two decades there has been a significant improvement in patient and graft survival in short-term. But there has not been a parallel improvement of patient and graft survival in the short term but there has not been a parallel improvement in the long term outcome. So this is of great concern of course. It is primarily based on 2 key publications based on data from the United States published in the American Journal of transplantation.

Slide 3

Doctor Meier-Kriesche is the first author in both of these papers and this one was published in 2004 and it says lack of improvement in renal allograft survival is despite a marked decrease in acute rejection rates of the most recent era.

Slide 4

In a follow up paper he says long term renal allograft survival have we made significant progress or is it time to rethink our analytic and therapeutic strategies?

Slide 5

The main graph from that publication is shown here he analysed results of transplants performed from 1988 to 1995 and found there was very little evidence of improvement, very tiny improvement and very disappointing. So that is what made the whole world react and repeat his statements that there really has been no improvement in long term outcome only in short term outcome.

Slide 6

I’m sure you’re familiar with this if you’re familiar with the transplant literature. Now I cannot speak about the -- or the United States’ data. I can only speak about the Collaborative Transplant study data this is a study we initiated in 1982, so it has now been in operation for 27 years on the principals that it is a scientific study, voluntary and open to everyone.

Slide 7

Amazingly more than 400 transplant centres are currently participating in this voluntary effort. This shows you that during the last 15 years or so we have consistently received about 2000 new registrations year after year after year. So there’s been no decline or increase fairly stable in the number of participants. Most of these transplants were renal transplants, they’re shown in blue and then there were liver grafts in purple and cardiothoracic transplants in yellow but I will only speak about the kidney transplants of course.

Slide 8

We are proud of the excellent long term follow up that we have. At 5 years nearly 90% of the patients have complete follow up. At 10 years still 77%, again it’s not ideal but those who know about registries most registries don’t publish these data because this is actually rather exceptional and very good on a large international scale with more than 400 participants.

Slide 9

Just to give you an idea these are the cities that are represented in the Kidney Transplant study. It’s a wide selection 46 different countries are represented and at the bottom you can see that we have more than 300.000 kidney transplants on record. So that’s the basis of my presentation.

Slide 10

Now if you take the first deceased kidney donor transplants that were performed from 1985 to 2008 and report it to the CTS study and you calculate the graft survival rate this is the line that you get. It represents 212.000 kidney transplants on one line and what you see is that initially there’s a relatively steep drop off in that curve so there are more failures in the early period and then it sort of levels off and there’s a continuing graft loss rate. This has been known since the early 1970s. We published in the early and mid 70s papers on the principals of this and I would like to repeat this here briefly because I think many people have not seen that or realised or digested that kind of a thing, also not the authors of the papers that I mentioned before. The first year is different. During the first year the loss rate changes all the time, it becomes smaller almost every week. In the first week is higher and the second week it’s somewhat less and so on. After the first year you reach an equilibrium around 1 year, not exactly of course. So that the one year point is sort of a magic thing. Only after that is the rate fairly constant.

Slide 11

If you plot the individual graft survival rates year after year, as shown here in dots and you plot them on a semi logarithmic scale, the left scale is now logarithmic, you can connect these dots with a straight line. That’s a very special thing because a straight line on this kind of graph means a constant rate of risk. If the risk is constant, you can calculate a meaningful half-life time and that is what has been done here. We say the half life time for first deceased donor kidney transplants done from 1985-2008 is 11.6 years. That means that at any point in time on that curve after 1 year, after 2 years, after 5 years, after 7 years if you take 100 transplants, it will take 11.6 years until 50 of those transplants will be lost or will not function any more. That’s the principal of the half life time but it’s only meaningful to talk about half life when the half life is constant. Now the attractive thing about this is if you have do left hand calculations you can look in the future if they are correct. You don’t need to do 10 years, you can take just the first 3 years connect the lines and make a 10 year projection and know what the 10 year outcome will be of transplants after 3 years of observation.

Slide 12

If we take those same transplants only calculate 4-5 years and extend the line to 10 and look at the angle of decline at the half life

Slide 13

and do a 10-year projection, then we obtain this here and it says that the 10-year estimate is 53%.

Slide 14

So we are slightly too optimistic here and that has been criticised and it is a well recognised problem. You could do a lot of acrobatics in statistical terms and adjust for that but it’s not really necessary, you know that you are slightly too optimistic but only slightly.

Slide 15

So we have two percentage points off by not going all the 10 years and using 5 years as an end point.

Slide 16

Why is that? It is because this is the 5-year period that we have used and then the extended line and the dots that we’re actually measuring and you can see that they drop off to the lower end so there’s an overestimation in the curve. The reason is the patient survival rate because if you think about it patient survival can never be constant. There is no constant patient survival, the older the people get, the more likely they will die. As the patients get older the patient death rate increases so patient survival cannot be a constant. There cannot be a constant half life of patient survival. If you deduct patient death and do what is called a death censored analysis, you don’t count patients who die with a functioning graft then you get a perfectly straight line or nearly perfectly straight line.

Slide 17

So if you death censored then it’s perfect if you count all cases and include death then it’s not perfect but almost perfect, just a slight deviation. If we do all transplants from 1985 on and subdivide them into 3 year intervals, you can see that starting at 1985 at the bottom with every 3-year interval there has been an increase and the question is, is that increase only due to early graft improvement or is the long term success rate also improved? For that we need the half life time and we need the log scale otherwise we can’t do this appropriately. If we put these same data on log scale and again based on a 5-year cut off and the dots are the ones that are actually measured after that projection, you can again see that the dots are slightly lower than the curves but only slightly but there has been clear improvement over time. So the 10-year estimate of survival has improved from 41% 20 years ago to now 65% there has been a fantastic improvement period after period after period.

Slide 18

If you do it in a death censored manner, it’s perfect. So clearly this improvement has been proven, there’s no doubt about it.

Slide 19

I would say that of all transplants being performed today a kidney transplant patient receives a cadaver kidney transplant, deceased donor transplant the likelihood that that transplant will function ten years from now is about 65%. 2/3 of the transplants will function 10 years from now of patients who are transplanted today.

Slide 20

You can do this in a more elaborate manner, you can do what you like, you can do a Cox regression analysis about 20 factors were -- into this here and you can compare each 3 year interval with the next 3 year interval. Like a top line shows if you compare 1998 with 1990, whereas 1985 - 1987 the previous 3 years, the hazard ratio of failure was 0.85 so there was a 15% drop in risk of failure. In the next 3 years it was 0.9, in the next 3 years it was 0.87 and so on. So in general about a 10% decrease in risk of failure in the 3-year intervals that we analysed. It’s very consistent and very interesting that this improvement has occurred over time.

Slide 21

Why has this improvement occurred?

Slide 22

Of course, everybody in this room knows because new immunosuppressive drugs came on the market. Because new drugs are much better than the old drugs and that is why the results have improved. Indeed if we look at the type of immunosuppressive regimens that were applied to patients, you can see in blue that tacrolimus came on the market about 10 years ago and has now basically taken over. More than 2/3 of the patients receive tacrolimus today. Azathioprine has been wiped out, has been replaced nearly entirely by mycophenolates. So the drug regimens have dramatically changed over time.

Slide 23

Is that the reason why the results have improved? The answer is no because if you compare the outcome of transplants treated with these various combinations of drugs, the results are exactly the same. So it is not the choice of the drug regimen per se, it must be perhaps better utilisation of the drugs. Again knowledge with experience but not whether you choose this drug or this drug. The result the outcome is really the same if it’s done properly. Moreover the long term advantage of mycophenolates over azathioprine remains to be proven. In fact, I think the evidence shows there is no difference.

Slide 24

Calcineurin inhibitor treated patients who were either on azathioprine or mycophenolate have exactly the same long term outcome, this here is started at 6 months post-transplant and long term after 5 years exactly the same curves as you can see. There’s no advantage that we can identify.

Slide 25

So what is it then? In my contention list there are many other factors that contribute to the overall improvement over time in transplantation and it’s to the credit of all of you that this has been happening. I’ll just point out a few examples of factors.

Slide 26

Let me point out the CMV prophylaxis in patients where the donor is CMV positive, the recipient is CMV negative. Clearly graft survival is better if the patients receive CMV prophylaxis you would say that that is clear, of course obvious

Slide 27

and the reason is this is death due to infection that patients who receive prophylaxis for CMV in that combination have a lower death rate due to infection.

Slide 28

Now over the years this has not always been clearly understood because 20 years ago only 30% of the patients received CMV prophylaxis but today 90% of the patients receive CMV prophylaxis and you can see there’s still a little way to go, those 10% also need to be changed of course, but this has been changing over time.

Slide 29

Total cholesterol, serum cholesterol the extremely high values are associated with impaired graft outcome.

Slide 30

This time it is not infection, this time it is cardiovascular death. The rate of cardiovascular death increases with rising total serum cholesterol.

Slide 31

There has been a dramatic change over the years in serum cholesterol. The low serum cholesterol fraction has increased, the high serum cholesterol fraction has decreased. So patients are much more favourable today than they were of course because of the introduction of statins etc.

Slide 32

Post transplant hypertension has been recognised as an important problem. This is a publication by us that we did together with the president of this society Eberhard Ritz in 1998 and we showed that the degree of hypertension was directly correlated with subsequent long term outcome.

Slide 33

Now interestingly if the participants in our collaborative study the ones who delivered this data they recognised this and at the time of publication when we worked on the paper right afterwards the blood pressure, the mean pressure dropped by 5 ml and another 5 ml the years thereafter and then stabilised again.

Slide 34

So we have achieved a drop in systolic pressure in patients under 1 year post-transplant. One can measure this at various time points of course but that is shown here and we were able to show that indeed this drop resulted in better long term outcome.

Slide 35

The red curve shows patients whose blood pressure was not lowered whereas the black line shows patients whose blood pressure was lowered from over 140 mmHg to lower than 140 mmHg we can take a certain cut off of course.

Slide 36

This also was published in great detail in a paper published in the American Journal of Transplantation.

Slide 37

So there are all kinds of things that have changed. Serum creatinine values have changed. The fraction of patients with good creatinine values shown in black has been increasing all the time with poor creatinine shown at the top in blue that has been decreasing all the time.

Slide 38

That of course is very important because function is the best indicator of long term outcome. This is the long term outcome depending on 1 year serum creatinine. You can see how dramatic these differences are.

Slide 39

Not only the one year but also the 5-year serum creatinines have changed, not as impressively but they have changed significantly one can calculate this of course.

Slide 40

So there’s less long term toxicity, better function in the long term why? Because long term drug dosages, toxicity has been reduced by dosing the drugs at lower dosages shown for cyclosporine at the left top, tacrolimus right top, steroids left bottom, MMF right bottom. In all of these statistically highly significant correlations, decrease of drug usage over time in recent years, much lower doses than in previous years.

Slide 41

But we also have pointed out that this is a 2-edged sword because if you go too low below certain thresholds and this also has been published in detail I can’t repeat it here or if you delete, withdraw these drugs completely, there is a risk of graft loss and a sizeable fraction of patients tend to loose their transplants if the drugs are decreased to a too low threshold, below a certain threshold or withdrawn completely.

Slide 42

Another point kidney preservation from a recent publication. Preservation times have decreased.

Slide 43

Cold ischemia times were much longer 20 years ago than they are today and of course, this is a good thing because the long ischemia times are associated with inferior outcome

Slide 44

but I’d like to point out also that up to 18 hours one cannot show any difference, there’s no gain from decreasing the cold ischemia below 18 hours for some reason with current preservation technology. But to eliminate those long terms makes a lot of sense and that has happened over time.

Slide 45

Many, many kidneys nowadays are transplanted with less than 18 hours or shorter cold ischemia and the greater than 36 hours can be -- has completely disappeared. It all makes good sense and that is why in total the results have improved and that is what I’d like to maintain here.

Slide 46

There are more factors and I’m not going to go into all of them. I could count up a few more. I would also like to say that there are factors that are difficult to measure with the approach that we are taking. For example, improvements in the diagnosis and treatment of infection. The improvements in the adjustment of immunosuppression. When you go lower, when you go higher, when you switch from one drug to another which of course, makes a lot of sense and you learn over time how to do that better. Improvements in general patient care including improvements in infrastructure at the transplant units. All that is difficult to grasp and difficult to measure but it probably also has contributed to the overall survival.

Slide 47

My conclusion then is that there has been a phenomenal very strong improvement in the success rate of deceased donor kidney transplantation over time, not only in the short term but also in the long term success rates. The long term probably chronic rejection rates or loss rates altogether have decreased, the patients are managed better today and the success rates are much better than they used to be. Today we can expect that 2/3 of the deceased donor transplants that are being done today will function 10 years from now. Thank you very much.

Slide 48

Chairman: Doctor Opelz thank you very much for your presentation. It’s a positive conclusion and are there questions? I imagine... yes please Claudio.

Question: Thank you Gerard for this beautiful presentation. Thank you for confirming that the results in the long term actually have improved. This is the general impression from the single transplant physicians in Europe. You pointed out rightly that it is important the so-called supported treatment in order to avoid hypercholesterolemia, hypertension and so on. May I ask you what is your impression about the use of corticosteroids or the avoidance of corticosteroids in order to further improve not only the graft half life but also the quality of life of the patient?

Prof. Opelz: Yes I didn’t have time to put all this into my talk, in fact we have conducted the largest ever conducted prospective study on steroid withdrawal. It was not randomised but prospective in kidney and heart transplant patients and pointed out significant advantages of steroid withdrawal in transplant patients. I think, as you are well aware, you can’t do this in every single patient but you can do it in patients who do well and in those patients it is to the advantage of the patient, better graft and patient survival and morbidity associated with corticosteroid use can be reduced. So there is no question and I think there we have an area where a lot more improvement can be achieved in the future. There are several areas where we can say that corticosteroid use is certainly one of them.

Chairman: Well, we are a little bit late I’m very sorry but there’s the next question please.

Question: Hello. Thanks Doctor Opelz for that nice lecture. I’d just like to ask one question about the difference between the azathioprine and MMF in the long term graft survival. There are data saying, actually pointing out that MMF has some superiority over azathioprine in the long-term graft survival but as you showed us there seems to be no difference what’s your opinion regarding that?

Prof. Opelz: I’ve seen the same papers as you. Maybe I should point out that this study is a rare example of a study that is not sponsored by the pharmaceutical industry. It is not supported in a significant way by the pharmaceutical industry. You show me the pharmaceutical industry independent studies on drug usage and then we can continue the discussion.

Question: --------

Prof. Opelz: No I’m not saying that, I’ve never said that. I think it’s wonderful that we have more than one drug, that we can switch from one drug to another when we run into problems. There is no question about that and the calcineurin inhibitors is in the same way. I am not against any drug I’m just saying it’s not as simple as that to say ok we buy drug A and drug B and drug A is better than drug B or B is better than A and we get 10% better survival. That you cannot expect.

Question: I agree but some people look at the cost of the drug because azathioprine is much cheaper than MMF.

Prof. Opelz: We have just published a paper in Transplantation on side effects associated with those two drugs for example and there are pluses and minuses on each side.

Chairman: Doctor Opelz we could follow and listen to you for another hour but we have to go on and it’s a real pity but I feel it would be very important that the public knew that we were so successful. Thank you very much.