Loading summary
Transcript23 lines
- [00:02]
A
Welcome to the American Society of Hematology Conversations with Blood Authors Podcast. This blood podcast episode is hosted by Dr. James Griffin from the Dana Farber Cancer Institute in Boston. He discusses ferric carboxy maltose increases fracture risk in patients and reduces bone formation in mice with iron deficiency anemia with Dr. Heinz Soler. He also discusses a low versus standard dose regimen as induction for AML, a multicenter randomized non inferiority trial with Dr. Raul Ribeiro.
- [00:35]
B
Hello everyone, my name is Jim Griffin. I am one of the associate editors of Blood for the podcast. We have two terrific papers to discuss today. The first is entitled Ferric Carboxymaltose Increases Fracture Risk in Patients and Reduces Bone Formation in Mice with Iron Deficiency Anemia. And that paper is going to be discussed by Dr. Heinz Zoller. Dr. Zoller is from the Christian Doppler Laboratory for Iron and Phosphate Biology at the Medical University of Innsbruck in Innsbruck, Austria. Dr. Zoller, hello.
- [01:11]
C
I'm most grateful for the opportunity to discuss our findings here. Diving straight into the question that we wanted to address with this paper, which we are really very excited to see now in print, is the question whether and why patients who get different forms of intravenous iron formulations actually develop clinical complications from hypophosphatemia. And in order to understand that question, I would just like to briefly review the state of the art of intravenous iron therapy which has profoundly changed since the introduction of these high dose intravenous iron formulations which can be given rapidly. So ferric carboxymaltose was introduced in 2010 and really changed the way how we treated intravenous iron deficiency anemia because we were now able to administer high doses of iron at relatively short infusion times. What was noted initially was that about 50 to 70% of patients treated with this new formulation developed hypophosphatemia. And at the time this was thought to be a consequence of the high demands of the bone marrow for producing energy and phospholipids. And so the idea was that these high doses just induced or the active production of red blood cells in the state of iron deficiency, which was now corrected. And the question that was in the field for a very long time was whether these high rates of hypophosphatemia actually had clinical consequences. And that's where we started. In order to investigate the clinical consequences of hypophosphatemia, we interrogated our own hospital database and identified about 180 patients who have been treated with this drug. And as an appropriate control. We thought we would use another intravenous iron formulation by the name of ferric dirisomaltose, which came into the market slightly later, but which was also a very comparable drug. We were really surprised to see that even after adjusting for age, sex and other risk factors as underlying osteoporosis, the fracture incidence in patients treated with ferric carboxymaltose, as the title says, was twice as high in those patients treated with ferric carboxymaltose as opposed to those patients treated with ferric derisomaltose. And of course, this is a retrospective study with a fairly low number of patients. And we really wanted to find independent proof of that really surprising observation. That's why we interrogated the Trinetics database, which is probably the largest database of electronic health records worldwide. We only investigated U.S. american patients whose data were available within the Trinetics cohort, and really confirmed exactly what we found in our own hospital database, which is fracture incidence is twice as high in patients treated with ferric carboxy maltose as opposed to patients treated with ferric derisomaltose. As a next step, we really wanted to understand what the basic mechanism, underlying mechanism is that causes this high fracture rate. At the time when we started the investigations, we thought, and what we were expecting was that hypophosphatemia was the driving force. And to really look into the molecular mechanism, we set up a mouse model of iron deficiency anemia, where we put mice on iron deficient diet and further aggravated the anemia by phlebotomy, which resulted in quite profound iron deficiency anemia in these mice. And we then treated these mice with either of these high dose and very comparable intravenous iron formulations. What we found was that iron deficiency in its own really reduced bone mass. So this is a novel finding that the bone quality, especially of trabecular bone, was massively reduced in those mice who were iron deficient. But when we treated those mice with either drug, we saw the same thing as in patients. Both drugs result in a robust increase in hemoglobin. So iron deficiency anemia can be treated very well in this animal model. But the extent to which the trabecular bone recovers is much more pronounced in animals treated with ferric thyrisomaltose. In other words, ferric carboxymaltose really further seems to inhibit collagen formation. And that's what we could further confirm in a cell culture model independently of the mouse model. Bringing all the molecular data together, we could show that Independent of hypophosphatemia, because hypophosphatemia could not be reproduced in these animals. Iron deficiency has a negative effect on bone and ferric carboxymaltose treatment appears to further aggravate these negative effects of iron deficiency, because this drug specifically goes to the bone. So ferric carboxymaltose has very specific bone seeking properties, is taken up by osteoblasts, inhibits their differentiation and their bone formation. And that is the current working model. Why we do think that even a single treatment course of ferric carboxymaltose further increases fracture risk in patients.
- [07:01]
B
Thank you very much, Dr. Zoller. You mentioned briefly that many people feel that iron deficiency anemia itself causes some osteomalacia or osteopenia. Is the mechanism of that known?
- [07:14]
C
The most likely mechanism of that is that iron is an essential cofactor of collagen prolyl hydroxylases. So for the collagen fibrils to build up this proper triple helix structure extracellularly, proline residues need to be hydroxylated. And that is a result of the action of the enzyme prolyl hydroxylase, which has an active iron in its centers. What we could also show in the mouse and animal model is that specifically collagen production seems to be reduced in those developing osteocytes. I think this is probably the mechanism. And having said that, there is data recently published from a group in Denmark who have investigated 140,000 healthy subjects from the general population and showed a very nice U shaped relationship between iron status and fracture risk. So there is an association between iron overload, but also with iron deficiency and fracture risk. So it has negative effects on the bone.
- [08:20]
B
Your data in the mouse model are pretty compelling that the addition of ferric carboxy maltose to these mice when they have iron deficiency anemia really makes the bone disease considerably worse. Did you happen to look whether FCM damages bone health in the absence of anemia, or is it something that requires the underlying bone disease of iron deficiency anemia?
- [08:45]
C
This is a very interesting question. The short answer is that we have really tried to recapitulate the human situation very closely. These drugs are used in patients with underlying iron deficiency. And that's why for this study we really induced iron deficiency anemia. Also in our animal model. The data that has been published really focuses on the combined effects of iron deficiency and ferric carboxymaltose. But having said that, I would say we do see also a tendency towards an improvement in Bone also after ferric carboxymaltose treatment. But it seems as if there are two counteracting effects add action in ferric carboxymal dose. One, the reduced collagen production by the drug. But on the other hand, the iron deficiency appears to be also ameliorated by this drug. So two things going on at the same time. And that's why we think that the control arm with ferric di or isomaltose is so important and shows that you can get a better improvement if you use a drug that doesn't have these unwanted effects that we've just discussed.
- [09:52]
B
And one question about the patients that you studied. Was there any effect of age? Did older individuals, were they more likely to have FCM induced bone health problems? And how about children? Are they at risk for this as well as expected?
- [10:08]
C
We found association between age and fracture risk. So older patients had a higher fracture risk. That's what you would expect. And we used that as an internal control, demonstrating the validity of both cohorts. But when we did a multivariate analysis for both cohorts, we could see that the effect of ferric carboxymaltose was independent of age and even underlying osteoporosis, which of course was the strongest risk factor apart from underlying kidney disease, which both increased fracture risk as well. But the effects of iron were independent of age. The second question is, is this of concern in children? Both drugs are being used and approved for being used in children, at least in Europe. And in a previous study we have shown that about 16% of children will develop hypophosphatemia. So in a way, children seem to be a bit protected from hypophosphatemia. We at this stage really do not know whether the negative effects that we have shown really also apply in children. And one could presume that growth may be impaired if it has negative effects on osteocyte development. But we would really prefer not to investigate this in a prospective clinical trial because we think that there is no ethical way to expose patients, and especially children or adolescents, to this additional risk.
- [11:43]
B
Thank you for that great discussion and for your really nice paper. We're going to go on to the second paper of the podcast today. The title of that paper is A Low versus Standard Dose Regimen as Induction for Pediatric aml. A multicenter randomized non inferiority trial that will be discussed by Dr. Raul Ribeiro from the Division of Leukemia Lymphoma, the Department of oncology at St. Jude's Children's Research Hospital in Memphis, Tennessee Dr. Ribeiro,
- [12:16]
D
thank you for the opportunity. Thank you. For the American Society of Hematology to allow us to present our work in this format. I also like to recognize and I am on behalf of my colleagues in China, including The first author, that is Dr. Li Gao and Dr. Who, a senior author for whom we are collaborating for more than 15 years now. This study initiated in 2012. Before that, Dr. Hu had come to St. Jude and visit with me and discuss opportunities to improve the treatment of acute myeloid leukemia in children. There were two main problems, the rates of abandonment. That means children would initiate the treatment and then the parents would refuse further treatment because of the intensity and the complications associated with the treatment and the cost. The second was the relapse rate. Mortality was also very high, so that the overall survival for kids at that time was like 35%. So discussing with her here at St. Jude, I offered to help her to develop the AML02 protocol that in our hands had been the best study here to that point. And a year later I visit with her and in fact the protocol that I recommended was considered too toxic for them. So it still had very high mortality rate, high refusal rate. So Dr. Hu, who also had education training in adult hematology, she told me that in Asia was very common for elderly AML to be treated with a very low dose cytarabine, half dose of intracycline and gcsf. And that was observed to have relatively good response rate in that population with 50% of the patients attained remission. I was not aware of that particular experience, but I reviewed the literature and decided to initiate that for the kids as a pilot. The first nine patients were really very impressed that eight patients had attained remission. And the doses of cytarabine is like homeopathic, 10 milligrams per meter square, subicutaneous by DID. And the anthracycline was half the dose. So it was 10 times less anthracycline, 2 times less in the GCSF. Initially I was a little bit perceptible about this because my experience with Central America, we were also developing treatment there. We reduced the dose of anthracycline and that combination reduced the CR rate was really low, relatively speaking, and the toxicity high. And it was expected that we could see something similar, but we didn't. Was actually cr. We extended for 40 some patients. Again, the same proportion of patients had attained remission. So then observing the experiencing elderly, we note that the patient received one or two courses and the disease became resistant and in the end would prolong survival but not eradicate the disease. We thought maybe then after the induction with low dose to decrease the rate of infection and also allow to educate the patients and the families and try to work together to establish resources and then treat after induction with exactly the same regimen that we use for standard approach to aml, including with palm air transplants. We extended this to include 80 or so patients present to the American Society of Hematology as an abstract ending by being published in Blood Advance. I remember very well that I submitted the paper to blood and Dr. Loewenberg say it's very impressive, but if it's not randomized, you're not getting blood. With that message we say, okay, let's do the randomized study. At that time more institutions were interested because of the success and we put together 10 institutions and randomized patients to receive the low dose regimen and the standard combination from the beginning and use a statistical approach that we want to consider known inferiority. Not exactly that we're going to ask it to be more effective. And that was what we demonstrate in this cohort that the patients who received low dose chemotherapy had no inferior treatment than the others that have the standard. And the overall treatment was very effective with international standards of 80% of the patients had overall survival.
- [18:24]
B
Thank you. Just a couple of questions. How many patients did you end up randomizing to each arm?
- [18:31]
D
200 patients in each arm. Although we continue after we determine the known inferiority, we continue with the same protocol to extend to include now more patients to see if we can have a more robust subtype analysis. For example, there are a signal that some types of AML would benefit more of the low dose, like kids under 3 years of age. But also that some patients with REC T1 with KIT mutation did not benefit. The standard was apparent to be superior. Now we have about 580 patients and we are analyzing now the subgroups to see if we can make more recommendations.
- [19:28]
B
A question about the role of gcsf. I believe you showed in your article that the low dose regimen patients received GCSF that was not part of the standard dose or higher dose chemotherapy regimen.
- [19:43]
D
Right.
- [19:44]
B
What role did the GCSF play here in getting such good results?
- [19:50]
D
Like I mentioned in the beginning, in Central America, reducing the chemotherapy to 35 milligrams per minute squared of anthracycline, three doses and 100 milligrams of cytarabine. The results were not impressive. They actually didn't attain remission most of them and a lot of toxicity GCSF to a two drug combination I think has a pivotal role in success of this combination. Compared what we saw in Central America and compared what we saw in Brazil. Now GCSF is well tolerated. There is no issue regarding to increase in white blood cell count and site reduction prior to initiate therapy for patients with more than 50,000 or 70,000 with site reduction with RSC, allow us to give the chemotherapy in Brazil. In addition to that, if the patient has 50,000, they start with low dose RSC, give one dose of anthracycline and then that is enough to site reduce and start the gcsf maybe third or fourth day. So it's very well tolerated. The mechanism in my view is that the patients with aml, they still have abundant normal bone marrow function. That the GCSF may have two mechanisms. I think in that situation. One is to protect the bone marrow environment to promote normal hematopoiesis and the other is that it may increase the sensitivity of the blasts to chemotherapy.
- [21:49]
B
Thank you. You mentioned that the CR rates were equivalent in both the low dose and the standard dose groups. And they were very high. Could you expand a little bit on long term survival? How long have these patients been followed now and is there any difference in long term survival?
- [22:06]
D
No, you're right. So CR rate was similar, no inferior MRD negativity. Early response was also comparable and long term survival. And then this protocol now has most of the patients have completed chemotherapy for more than two years, was measured in three years survival. There is no difference in long term relapses or differences that we can say in survival.
- [22:40]
B
So this was a study focused on pediatric patients. What are the implications of this for older patients for adults?
- [22:50]
D
That's interesting question and I always ask myself why in Asia this combination is really heavy utilized was initially in Japan in adults, was the very first study in Europe and the United States. My impression is that at the time this observation in Asia was done, it started in the United States the use of hypomethylated agents in adults with aml. And that was the line that the Americans decide to go or the Europeans as well. And initially with ARAC and the cyto being alone and then combined with low dose ROCK and eventually with Venetoclax. So it was a, I think a different pathway. And to my knowledge even the adults in Brazil now they are discussing low dose intensity, but they are opting for the venetoclax based on the paradigm study. And Ducitabine or azacytidine, I'm not sure if they think, well, 10 milligrams per meter square subcutaneous would not be appropriate for treatment of AML. I think there is this barrier that has to be crossed to us. Accept that.
- [24:19]
B
Thank you very much. The results of this study are really impressive and I certainly encourage all the listeners to get more details from the original BLOOD article. With that, we'll conclude our discussion of these two terrific papers. Thank you very much.
- [24:34]
A
Thank you for listening to this episode of Conversations with Blood Authors. To read the articles, visit bloodjournal.org this episode is copyrighted by the American Society of Hematology.