Part 1 - Comprehensive Diagnosis and Management of POEMS Syndrome

00:11 Stephen Hibbs: Welcome to this episode of the HemaSphere podcast. My name is Stephen Hibbs and I'm a hematologist and clinical research fellow based at Queen Mary University of London, and I'm one of the Scientific Editors for HemaSphere.

So today I’m joined by both a hematologist and a neurologist to discuss a condition that straddles several different specialities. Dr. Shirley D'Sa is a consultant hematologist and honorary Associate Professor based at University College London Hospital. Michael Lunn is Professor of Clinical Neurology, based at University College London and the National Hospital for Neurology at Queen Square.

They published a HemaSphere paper last year entitled "Comprehensive Diagnosis and Management of POEMS". And in this podcast, we will focus on some of the diagnostic challenges that POEMS presents. How do you discuss a complex condition like this in simple terms? What challenges and opportunities cross-speciality work it brings, and some of the uncertainties about the condition and the ways forward.

So the first question I wanted to ask you both - and I'll start with Shirley - is: How did you both end up being interested in such a rare condition like this?

01:14 Dr. Shirly D'Sa: Yes, thank you. So my background in hematology is myeloma-based. And during my registrar training and subsequently when I became a consultant, I was in a myeloma clinic, in proximity to Queen Square, having it in the same trust, meant that we already had some established collaborations.

My supervisor at the time, and our colleague, Professor Kwee Yong, was in communication with Professor Mary Reilly at Queen Square. There was the commencement of a joint clinic, to look at some of these patients.

So that's how it started, but I think, once Mike Lunn joined the team as a consultant on his side, we got into the clinic in a more systematic fashion. It became more frequent; more regular. And I think we both came to it from our different angles, but it is a fascinating disease, to start with. It is multi-systemic.

And my own personal journey is - although I'm a hematologist, I do love the physician in the aspect of hematology. You have to look at the patient as a whole. You have to understand all the nuances of the condition. I've always valued working very closely with Mike, because I see medicine through a different lens as well as my own. So I think it's been an organic process. It's grown.

But I think POEMS is such a unique and fascinating condition; and one in which you can see amazing change, if you pick the diagnosis up early and treat it effectively. So that, I think, is really the way I've got in.

03:08 Stephen Hibbs: How about you, Mike?

03:11 Prof. Michael Lunn: Yeah, I think that's very similar really, for me. It was this collaborative working that was always great fun. I took over from Mary Reilly in the joint clinic.

I used to sit in that clinic occasionally as a more junior doctor, and was fascinated by the complexity of some of the patients.

But I suppose, when I got my consultant job and had to start making my own competent decisions, the challenging patients are always the most fun and they're the most intellectually difficult - and the most satisfying to solve.

There was one in particular who I remember extremely well, who was young but very disabled over a short period of time, who had been through more than one other neurology centre and then came to us. We perhaps have, to an extent, the benefit of retrospect; in looking at everything else that everybody has done and failed in a way; which, I suppose, puts us at a bit of an advantage. It is much more difficult right at the beginning.

This patient was very complex, with systemic disease and with all sorts of interesting things - internal organs and skin and nerves. And in some ways it felt like a very wholesome disease that didn't look very mysterious from the outside, but actually when you looked into it you thought "How can all these things happen together?"

So it became a fascinating case in its own right. It was solved fairly straightforwardly, as it turned out, with a bit of additional thought. The patient was then fixed - and we now see him from being totally disabled. He still comes back to the clinic. It's enormously satisfying seeing him nearly twenty years later now being normal.

But it was then and a few other very similar cases where you think "Oh, I've seen this before"... and nobody else has seen it before because it's so rare. And actually now that we see many of these cases, despite their variability and differences, actually they look largely the same as one another.

And so, it's enormously satisfying to be able to put all the pieces of the jigsaw together and recognize all the different aspects of it and say "Okay, well that is what this is" and then have a treatable condition that people get better from. So it's a very satisfying place to be.

Of course, it's only one of the many diseases that we do deal with, but as we've developed a reputation and we've seen more and more patients, that we're able to give them more and more comprehensive therapy. We're extremely lucky being in a centre where we can go from the beginning to the end, from initial symptomatology through the diagnosis, through complex treatment to rehabilitation - and hopefully back to normal life.

So it feels like a very wholesome place to be and I think we both enjoy it. We enjoy each other's intellect and company at the same time. It's great.

06:32 Stephen Hibbs: Both of you actually alluded to both the importance and the challenges of diagnosis. You mentioned the big "if" of if people have been diagnosed early enough, in terms of the sort of recovery that can be expected. Mike, a few times you mentioned just how challenging this can be, particularly if you don't see it very often.

So I was struck in your paper, actually, by that message, of just how important and how common delayed diagnosis or misdiagnosis is. Mike, I'll start with you. Would you be able to kind of just say a little bit about what a journey to diagnosis might look like for a patient with POEMS and the different steps at which it might either be missed or caught?

07:13 Prof. Michael Lunn: Yes. So sadly the diagnosis is often slow and perhaps delayed. One of the interesting features of some of the data we've produced is that the average time of diagnosis was about 17 months; which is quite a long time and is now getting shorter.

The other thing that we discovered was about 50% of patients made their own diagnosis, or at least had suggested it. Quite often, it was rebuffed by the clinician to whom it was suggested... and that's because those patients; we've all, in medicine, become very focused on one disease or one area. We've become super-specialists.

And so, the patient journey - if the patient starts off with a neuropathy and numb/tingly feet or some pain - if it's painful, they might end up in a rheumatologist straight away perhaps; and then they're off down a wrong path. They might end up at a neurologist, but the neurologist has to think more broadly than the neuropathy.

They might appear randomly with a hematological problem and almost no neuropathy at all, because the P of POEMS is polyneuropathy, and the hematologist doesn't necessarily immediately think about it, if the patient doesn't complain of the polyneuropathy.

So at the early stage where symptoms are mild, you can be tripped up if you don't start thinking more broadly.

We did then publish some -- If the person gets to a neurologist -- We published a paper that suggested if you have a conduction slowing neuropathy, a demyelinating neuropathy; then you can really rapidly get to a diagnosis by doing a serum immunofixation. It's a very simple test on the blood.

And by measuring her (VEGF), her vascular endothelial growth factor; both of which only cost about £40 each, but can be used within a sort of series of blood tests that you would be using in your conduction slowing neuropathy blood set. And that would give you the diagnosis right at the beginning, because there's very little else that produces an abnormal and a light chain paraprotein and a very high VEGF (in the context of a conduction slowing neuropathy).

So that's a change in perhaps a diagnostic strategy that we need to try to introduce to neurologists - because they can now do a VEGF. And if neurologists can access that, which should be fairly easy, then they can make that leap early on without going through several other stages.

Perhaps I won't speak for Shirley's side of things in hematology and where the diagnosis gets missed there, but I think probably, if people miss those steps, they then get treated for a conduction slowing neuropathy because the neurophysiologist might say "Well, this is conduction slowing. It must be CIDP" and the patient gets steroids, which then squash the VEGF, and then it becomes more difficult to find or they get lots of IVIG and it takes several cycles (at six weeks each) to work out they haven't responded. So it might be six months before they haven't responded. Then somebody thinks again.

So I think the message is, for my neurology colleagues and perhaps my hematology colleagues referring to neurology, is: If you've got a neuropathy and it's a conduction slowing neuropathy, then just think "Could this be a bit more complicated than I think it is?" and do those tests.

I mean, again, going back to the retrospect thing... We have the advantage, by the time we see patients, perhaps say more like six months on now than seventeen months on. And that usually many of the other features are visible.

There's some ankle swelling in the context of a neuropathy. For a 40-year old for instance, would be really unusual. Not many 40-year olds have ankle swelling. So you know, that should ring a bell. And if you take somebody's shirt off - which many clinicians fail to do - then you might find the classical skin lesions.

You have to look at the fingernails. This is something that we're all taught in MRCP, you know? Are those fingernails normal or abnormal? Is the colour of the toes normal or is it acrocyanotic? We have to take the socks off.

So it's something that people perhaps have lost the skills of doing, or they're too busy to do in a busy clinic, but it is just thinking "Oh, I just need to go that extra step. Take the socks off. Take the shirt off. Look for the exciting features that give me all the jigsaw puzzle. Ah! Ping! Got it!"

So some of it is just fun general medicine, really. All of those things can trip you up. If you think of it, at the beginning, in many respects it doesn't matter if you sort of over-make the diagnosis. If you think about it and you can reject it, in a way it's better than if you haven't thought about it and you missed it. So I guess that's how I think things get missed or get picked up.

The other thing that makes people - thankfully - get to the diagnosis more quickly is being able to tell people that we'll diagnose this like this, and educate people. We try to give as many talks as we can. We publish useful papers like this one, that help people in their clinic, that they can reference whenever they want to. You know, patient information events. It does get -- The information gets around. So it gets in peoples' psyche and they do think about it.

13:14 Stephen Hibbs: Brilliant. So Shirley, what about some people coming through the hematology clinic or I guess where there's the monochromal gammopathies being picked up? Perhaps the other features haven't yet been noticed. What might a journey look like through the hematology clinic?

13:31 Dr. Shirley D'Sa: Yes. I think what Mike and I have found over the years is that quite often POEMS is hidden in plain sight, actually. And because we've seen many, many patients over many years, we have just become adept at spotting the features, you know?

And so, in publishing our paper; our various papers, but most recently the Comprehensive Diagnosis paper, we were hoping to just draw out those features so that people spot them. And I think in hematological circles; the monochromal gammopathy, which is LAMBDA-restricted primarily (almost all are LAMBDA-restricted) with myeloproliferation - so people have a high hemoglobin and high platelet - one should really try and think about this.

Now I think, as Mike alluded to, modern medicine sort of really disadvantages the POEMS patient, because people are much more specialized. So even in hematology... I mean, I think I'm speaking from the perspective of a specialist hospital. So we have people who specialize in MPN. We have people who are plasma cell people, etcetera.

I mean, most hospitals will have a hematologist who will see high platelets or arythrocytocis and think "Okay, is this an MPN? Should we do JAK2?" etcetera. But I think people tend to compartmentalize these different problems. So someone who is dealing with an MPN-type thing will very rarely even think about doing a protein electrophoresis because it just doesn't enter their sphere, you know?

I think that is - if we're going to get all philosophical about it -- I think the way medicine is practised nowadays is sad, actually. I think there are many plus points of being very specialist, because if you have a problem that requires hyper-specialisation, then you're in the money, you know? But if you have multi-system problems, you need someone to join the dots... and I think that is where people can fall between the cracks.

And that is why Mike -- I mean, we have been at pains, as Mike says, to give talks and to publish things that will be helpful; to make it easy for people to pick it up. I have to say, as the more we've done over the years, the more people do get in touch, and say "Oh, could this be POEMS?" And it's always better to say: "It could be, but it isn't". In fact, this week alone we've had two or three –

-- Well, I gave a talk last week in the Midlands about POEMS. People... They think about it. They think "Oh, my gosh! Have I missed POEMS along the way?" We've had some emails and, you know, that is great! Because, if we –

-- What I always say to people is, when we think someone may have POEMS: In our mind, it's like almost an emergency, a medical emergency, because every week that you delay the diagnosis usually means months of rehab for the patient at the other end. Plus or minus worsening performance status, because they often have cardiorespiratory compromise and other things, which means more therapy to get them optimised etcetera.

16:53 Stephen Hibbs: Just to clarify one particular thing with this, I guess. You're kind of in a -- I don't know -- MGUS clinic or something else where someone might be coming through, and you think about POEMS and you do a VEGF - and it's normal. Is that -- How unlikely does that make it?

Is a sort of "normal" VEGF really almost incompatible with -- In a patient not on steroids. Is that almost incompatible with POEMS, or are there some kind of other things that you've got to think about, apart from the steroid; the IMEC steroids here?

17:27 Dr. Shirley D'Sa: Mike, do you want to go for that one?

17:30 Prof. Michael Lunn: Yeah. I mean, we haven't seen exactly -- With those caveats, we haven't seen many patients with POEMS who have a normal VEGF at presentation. I think the VEGF is definitely squashed by steroids.

But otherwise, it would be unusual. The VEGF is usually very high, extremely high. I mean, at levels that you don't see with any other disease, because some other things do push the VEGF up a bit. Iron deficiency, hypoxia disorders and "Oh well, it's just COPD or sleep apnea". We've had a few patients who've had Methemoglobinemia who have pushed their VEGF up as well.

18:10 Stephen Hibbs: Yes.

18:11 Prof. Michael Lunn: You know, so there are things that make it go up, but very little keeps it suppressed.

18:15Stephen Hibbs: Yes.

18:16 Prof. Michael Lunn: So it is only though, in the criteria, you have to have a polyneuropathy and you have to have a monoclonal disorder; whether that's a solitary plasmacytoma (which may not be secretory) or whether it's the evidence of that secretory monoclonal disorder in the blood, for instance.

So you have to have the polyneuropathy in that, and then the VEGF is not necessary. But in practical terms, it's very seldom to see a patient at presentation who hasn't got a VEGF that is exceedingly high.

18:54 Stephen Hibbs: Yeah. Okay. No, that's useful.

18:56 Prof. Michael Lunn: So it's a really useful biomarker. I mean, it's unusual in biomarker terms, actually. It is not specific…

19:05 Stephen Hibbs: Yes.

19:06 Prof. Michael Lunn: But in the right context, it is hugely sensitive. You have to have it in the right context. You can't just say "Oh, a patient comes to the door. Test the VEGF. Is it high? Got POEMS. Is it low? Hasn't got POEMS". It's part of the jigsaw puzzle, but it is an exceptionally good biomarker.

19:23 Dr. Shirley D'Sa: One point I just wanted to add was that, in terms of the polyneuropathy, we have seen occasional patients with Castleman variant of POEMS who have very little in the way of polyneuropathy, if anything actually. Isn't that the case, Mike?

19:37 Prof. Michael Lunn: Yes. Those Castleman's patients have almost no neuropathy at all, but they might have symptoms that sound polyneuropathic. We've certainly had patients who've --

-- One particular patient who saw Mary Reilly; ten years before we subsequently saw them, who -- I still have the letters of that interaction, where the symptoms sounded like polyneuropathy, but Mary could find no evidence of it clinically or electrophysiologically. And ten years later, came back with a much more florid picture, and a clear polyneuropathy with it.

But Castleman's tends to have very little or no neuropathy, but I think, Shirley, I would say that (in every case of Castleman we've seen) the VEGF is sky high.

20:24 Dr. Shirley D'Sa: Absolutely. Absolutely. In fact, recently at the meeting I was at, the team there presented their case of a patient who had everything but neuropathy - and they were really quite befuddled by that, which is fair enough, because P for POEMS etcetera. But actually, that is one scenario where you don't see much polyneuropathy.

And an important point, while I'm on the subject, is that such patients, even though they have so-called "Castleman variant", they should not be treated with drugs like Siltuximab, because they need POEMS treatment. That's another thing. We have seen a few patients who have, because of the Castleman finding in their nodes or whatever, they've gone on to have Siltuximab, and that's actually just delayed the treatment for the neuropathy, which is anti-plasma cell.

21:15 Prof. Michael Lunn: I think going back further on from that there's always these extra little tips, aren't there? POEMS doesn't have to be POEMS, in terms of all of those features.

21: 26 Stephen Hibbs: Right.

21: 27 Prof. Michael Lunn: You certainly need enough of them, and in general, you would definitely like to see a "P" and an "M". That's pretty crucial, except in that Castleman's circumstance.

But in the Castleman's circumstance, which would have a VEGF that was very high just because it was Castleman, you then need some of the other features in order to get at least an "OEMS" or a "PMS". You'd need at least three of those features to really make yourself comfortable. And of course, in the diagnostic criteria there are many more; and actually, when you look, you actually find you've got lots more bits of the jigsaw.

So you even have to think beyond the box of POEMS --

22:07 Stephen Hibbs: Yes.

22:08 Prof. Michael Lunn: -- as an acronym. You have to think within the whole diagnostic criteria. There's a bit of boxing and coxing sometimes. Although, actually we don't ever find that we're hammering a case into the diagnosis; that patients fit themselves nicely into diagnosis nearly always, when you look hard enough.

22:35 Stephen Hibbs: So it turned out that we had more to talk about with Shirley and with Mike about POEMS than I'd initially anticipated. So we've decided to make this into a Part 1 and Part 2 will follow shortly. So please tune in soon for the second part of this podcast.

In the meantime, do have a look at their POEMS paper on the HemaSphere website, and we'll be back with Part 2 soon.