Slide 1

Good morning. Thank you kindly for the invitation to talk in this beautiful city and thank you to the audience for staying. I see three quarters have already gone, normally only 1% remains so thank you very much. I’m going to talk about work that has been done at the University Hospital in Birmingham in collaboration with GAMBRO and it’s about an important new dialysis membrane for helping patients with multiple myeloma and cast nephropathy.
First of all, I’d like to set the scene with the problem in multiple myeloma and then how we should measure free light chains and then the use of this dialyser.

Slide 2

The problem is that 50% of patients have light chain renal damage at some time during their myeloma disease. 12 - 20% of patients first present in acute renal failure from free light chain cast nephropathy. 10% remain dialysis dependent long-term. There is a high mortality rate. Chemotherapy and transplantation are difficult and hazardous because it’s not easy to set the drug regimens. Cost of haemodialysis is 70,000 dollars per year and there are approximately 50,000 new patients with renal failure and myeloma worldwide per year.

Slide 3

So, the first question is how to measure free light chains in these patients. Do we measure urine Bence-Jones protein or serum free light chain immunoassays?

Slide 4

So, we need to look at some basic physiology of the kidney. In the nephrons we have 1 million nephrons the light chain can enter the glomerular fenestrations because of its small size. Kappa light chain is typically 25 kDa as a monomer, whereas lambda light chain is typically a dimer at 50 kDa but this rapidly passes into the glomerulus with a half life of kappa of about 2 hours and the lambda dimer of about 5 hours. The production rate is about half a gram a day, so these filtered light chains enter the reabsorption mechanism of the proximal tubule which is highly efficient and it can remove 10-30 g/day, maybe 20-30 times the normal production rate. There is always a small amount of free light chains in the urine. There is a mucosal surface on the distal part of the nephron and in the urethra this produces large immunoglobulin molecules such as secretory IgA in the mucosal surface and also a little free light chain from that mucosal surface. This enters the urine and can be detectable by sensitive assays but we can be sure that it’s impossible if the kidney is normal for this free light chain in the serum to pass through the proximal tubule because of the huge reabsorption capacity. This means that urine tests for free light chain are very insensitive for small production levels of free light chain and cannot be used.

Slide 5

We therefore, looked at the possibility of measuring serum free light chains. This had been considered for 30 years but there is a huge technical barrier. The serum contains intact immunoglobulin, this is the heavy chain and it contains bound light chains. These are the free light chains; kappa monomer and lambda dimer and there is a surface that is hidden in the light chain, in the intact immunoglobulin which is exposed when the light chain is free. This surface is about 5% of the area of the molecule and then antibody assay must be targeted against this small surface area because if the antibody binds to any other surface component of the light chain, it will cross react with the light chain on the intact immunoglobulin and give a false high result.
Since there is a thousand times more light chains bound in the intact immunoglobulin in serum than free light chains, any assay must be extremely specific.
The technical breakthrough was 6 years ago when we developed knockout sheep. The sheep are knocked out against intact immunoglobulin surface epitopes and then we could immunize these sheep with the free light chains and they only make antibodies against the hidden determinants and it is this technique which has allowed the possibility of serum free light chain assays.

Slide 6

If we then compare this immunoassay for free light chains with the normal laboratory methods, we can see that this is the upper limit of the normal range of serum free light chains, about 20 mg/ L and these are the light chain concentrations for comparison against the existing laboratory tests. Serum protein electrophoresis cannot detect patients’ serum free light chains less than 1000 mg/L. Immunofixation electrophoresis is better, capillary zone electrophoresis and total kappa and lambda only measure the intact immunoglobulin.

Slide 7

Here is the free light chain immunoassay. For the first time we can measure the light chains in a normal range and it allows detection of all the patients in this area. This is a revolution in sensitivity and we can compare it with pregnancy tests. Imagine if we had a pregnancy test that missed 30% of the pregnancies, we would have some very unhappy ladies. So, now we can detect all these patients in the serum. This is urine protein electrophoresis, it is good sensitivity because we can concentrate the urine 100 or 200 times but as I indicated, the urine is bad because the kidney will not leak free light chains, if the function is normal.

Slide 8

If we now look at these results compared with a patient, here is the free light chain concentration in the serum, here’s the serum normal range, here are the serum concentrations. As the tumour free light chain myeloma tumour increases in size, the serum concentration increases in parallel. Whereas the urine concentration remains very low until there is overflow in proteinuria. Later as the kidney is damaged by the free light chains entering the distal tubule, the serum free light chain levels rise much quicker because of the extended half life of the serum light chains from renal failure, whether it’s complete renal failure the serum half life of light chain is 3 days up from 3 hours and the removal is by the liver and other reticuloendothelial cells. In contrast, the urine levels reach a peak and when there is progressive renal damage, there is a fall in urine free light chains and when the patient has no renal function, there is no light chain in the urine. So therefore, it’s impossible to monitor patients properly from urine measurements of light chains if they have renal failure.

Slide 9

So, this is a patient who is being monitored for myeloma, it is IgA myeloma with lambda serum free light chains. This is the concentration of creatinine that you can see is normal but there is a very high concentration of serum free light chains which maybe normal in the urine if the kidney is functioning normally. We can then see a rise in the free light chain in the serum with a small rise in creatinine and this is before the patient presents in acute renal failure. The patient is given chemotherapy --- in dexamethasone and we see a fall in the free light chain to normal with good chemotherapy but the kidney does not recover. So all these patients should be monitored with a serum measurement to prevent these patients developing relapsing renal failure. The mechanism of course, is that the light chains pass the proximal tubule, they enter the distal tubule and they combine to form a cast nephropathy with Tamm-Horsfall protein.

Slide 10

This is a urine cast and we can stain them in the distal tubules with histological techniques.

Slide 11

The course of these high serum free light chains, cast nephropathy has been treated by plasma exchange to remove the serum free light chain levels.

Slide 12

However, more than 80% of free light chains are extravascular, the distribution volume in the body is similar to creatinine. We would never think of removing creatinine by plasma exchange, we always use dialysis. It’s ridiculous to use plasma exchange in these patients. The plasma exchange procedures are of limited duration 2 weeks in frequency, maybe 6 plasma exchange volumes are removed. Current chemotherapy may not stop free light chain production for months, longer than plasma exchange procedures of 2 weeks and longer than renal recovery times from cast nephropathy and the typical recovery rates are only 10-20%.

Slide 13

If we look at a model of the light chains in the body, we have a serum compartment, a vascular compartment and an extravascular compartment. The free light chain enters the vascular compartment from the bone marrow, from the tumour and it leaves by the kidney or we can remove it by dialysis or plasma exchange or it gets removed by the reticuloendothelial system. But 80% of the light chain circulates in the extravascular compartment with a forward rate and a return rate. Some can be removed by the reticuloendothelial system in the extravascular compartment. If we put the figures, the numbers of light chain concentrations in patients into this model, we can predict the effectiveness of plasma exchange and haemodialysis.

Slide 14

This is a model, it’s a patient who presents with 10,000 mg/L of serum kappa, a typical concentration of a patient with myeloma and renal failure and this is the half life of clearance in days for the free light chains, if we had chemotherapy that would immediately stop production. More typically the chemotherapy kills about 10% per day of the myeloma tumour cells. Then this is the half life curve for the free light chains with this effective chemotherapy. You can see the concentrations remain high for 25 days. The normal concentrations are 20 mg/L, at 500 mg/L we get renal damage. So, the kidney will continue to have damage from the light chains in the blood with good chemotherapy for at least 4 weeks and this is probably longer than the recovery time for the kidney in these patients. This is the effect of plasma exchange, this is the one plasma volume removal, recovery, 2, 3, low week end, 4, 5, 6, and you can see that 6 plasma exchanges only removes about 20% of the free light chain in this patient and we still have very high levels at the same period at 25 days.

Slide 15

When we made these calculations, we realised that plasma exchange is unrealistic so we looked around the world to all the different dialysis manufacturers for a membrane, a dialysis membrane which would remove free light chains. The normal pore size is about 10-15 kDa. Since the light chains are up to 50 kDa we must have a very big membrane. On a shelf in GAMRBO we found a protein-leaking dialyser with a high cut-off of nearly 60 kDa, similar to a normal glomerulus. So 1.1 m² dialyser and this we compared with all other dialysers and this was easily the best.

Slide 16

We tried in the laboratory simulating removal using this membrane. This is a 4-hour haemodialysis on a dialysis machine with free light chains and you can see because of this 4-hour period of dialysis compared with only 1.5 hours for the plasma exchange, we naturally get a far greater removal from the extravascular compartment. Then we can perform the 4-hour dialysis 3 times a week, as many times as possible and now we have a significant removal of free light chains at 20 or even 15 days. We can extend the haemodialysis. This is 12 hours of haemodialysis every day in these patients and now by day 5 we can remove 90% of the free light chains.

Slide 17

So, with this information we tried a patient who had chronic renal failure and relapsing myeloma. This is a 6-hour, a 300 minute dialysis in a patient with serum free lambdas with about 9000 mg/L. This patient was in chronic renal failure, so we expect no recovery but it shows that we can reduce the serum light chain concentration by 50% in 6 hours. But even more important this is the free light chains in the dialysate fluid and we’re removing about 200 mg/L, so in a 200L dialysis we can remove 40 g of free light chains from this patient. With this information we then waited in the hospital for a patient. Then the luckiest man alive arrived from the airport. He was a Birmingham person, he had emigrated to Jamaica when he retired and after a few weeks in Jamaica he went to the Kingston hospital with bone pains. They told him he had myeloma and he was in acute renal failure. They do not treat these patients in Jamaica and that he would die within 2 weeks. Because of his connection with Birmingham he phoned up British Airways, he got a one way ticket to the airport in Birmingham. He then phoned the ambulance and got a one way ambulance ticket to the hospital and he arrived in the ward just as we were opening the membranes. This is what happened to him. Here are his kappa light chains, they are 1000 mg/L. He was in complete renal failure, this is a very toxic light chain. He had dexamethasone, post and this is the removal of the free light chains during the dialysis period.

Slide 18

So this is a 6-hour dialysis, we removed 3.2 g and here’s another dialysis period, this is no dialysis and you can see the fall in the free light chains, then the rise between chemotherapy and then here there is a fall in the free light chain between dialysis periods because the kidney is recovering and metabolising the free light chains. By 4 weeks the patient was off dialysis.

Slide 19

This is now 18 months ago, the patient’s had a bone marrow transplant, the GFR is in the order of 50 and the patient is in complete normal health.

Slide 20

This is another patient, it’s a relapsing myeloma that the patient had failed the first course of chemotherapy, cyclophosphamide, thalidomide and dexamethasone. This is the serum lambda concentration. You can see this is the dialysis, there is no removal of free light chain of significance. So we added bortezomid, VELCADE and now you can see the effect of one dose of bortezomib and now the reduction of free light chains is quite clear. This combination of the removal by the dialyser plus effective chemotherapy. This patient is in complete myeloma remission with good renal function. This patient failed the treatment. Here are the treatment days, here’s the kappa, it’s 50.000 mg/L, so fairly a non-toxic light chain because the levels are so high. The patient was only given dexamethasone. You can see reduction but always it rebounds back to this plateau level which only reflects the reduction because of the membrane. In the first ten days we removed 346 g of free light chain kappa, 242 g. We removed a total of nearly 2 kg of free light chains and there was no response because the patient became infected, we had to stop chemotherapy and inevitably the patient kidney did not recovered.

Slide 21

We treated 9 patients with acute renal failure and cast nephropathy by free light chain removal haemodialysis over 14 months. 7 of 9 were dialysis independent by 6 weeks and remained off dialysis. The median estimated GFR increased from less than 10 to greater than 40 with a good range. 3 of 9 patients are in complete myeloma remission following bone marrow transplant. 2 of 9 of patients had no renal recovery because of ineffective chemotherapy. The serum light chain measurements allow us on a daily basis to monitor the chemotherapy and the dialysis and we can change the chemotherapy on a daily basis if it is ineffective. 1 of 9 patients had impaired renal function before treatment and recovered, so this treatment should also work for relapsing myeloma. The cost estimates of this acute treatment over 4-6 weeks is much less per year than chronic haemodialysis.

Slide 22

Improved renal function, renal recovery using free light chain removal haemodialysis, this is the statistics, this is the dialysis independence, time in days. This is a control group from Birmingham for the last 6 years of 30 patients with myeloma and you can see that the renal recovery in these patients is normally complete within 6 weeks. These levels of renal recovery are much higher than any reported in the literature.

Slide 23

The survival, improved survival using free light chain removal haemodialysis, again the accumulative survival against the historical controls, there is no significance yet, we’ve had one death in this group and that was a patient who failed to get renal recovery and the other patients are at an early stage but normally we see early deaths in these patients and the typical survival in these patients who fail renal recovery is 50% at 6 months.

Slide 24

So in conclusion, the free light chain removal haemodialysis by using the GAMBRO high cut-off dialyser is highly effective. Patients may become dialysis independent over 4-6 weeks, effective chemotherapy is essential. There’s a long-term increased renal recovery in patient survival. Multicentre evaluation and clinical trials are underway throughout Europe. Thank you.