A Deep Dive Into Complex Clinical Scenarios Involving Refractory CMV Infections

[00:54:14]

Patient Case #2

The first case is a 64-year-old woman who had thin basement membrane disease, who underwent kidney transplant from a deceased donor, donor-positive/recipient-negative for CMV, of course.

The surgery was complicated by thrombotic microangiopathy from tacrolimus. So she was converted to belatacept/prednisone and azathioprine. And she didn't tolerate mycophenolate either.

She was on—so already you're kind of thinking oh, D plus or minus belatacept, okay. So 6 months of valganciclovir as per institutional protocol. And 1 month after the valganciclovir prophylaxis ended, she was on surveillance after prophylaxis. So we knew that this was a high-risk situation. So we were monitoring her weekly. She developed CMV.

And as you can see there, her initial CMV DNAemia was just over 10,000 IU per mL. She was started on valganciclovir treatment at 450 mg twice a day. Week 1, her DNAemia went from 10,000 up to 22,000. You know, it's not even half a log difference, but still it goes up at 1 week. Just like I said, commonly happens. No drama.

By the second week, she's down to 17,000. Okay. Kind of wish it was a little lower than that, but nonetheless—you know, anyway. And the following week, she's down to 1677. So great. Then 261 and then undetectable. And we switch her to prophylaxis.

Okay. So like a pretty good story overall, although you are kind of thinking, well, D plus or minus and belatacept and she's presenting this case, right?

[00:55:50]

Poll 5

So she's on prophylaxis. And while on prophylaxis, she has breakthrough CMV DNAemia at 18,000 IU per ml on plasma. So what would you - what would everyone in the room do and everyone online, what would you do at this point? Would you:

1. Start maribavir;
2. Start foscarnet;
3. Start letermovir;
4. Start higher dose valganciclovir treatment;
5. Start intravenous ganciclovir treatment; or
6. Would you wait for resistance testing to return in like 1 to 2 weeks.

So go ahead and vote, and let us know what you would do.

[00:56:41]

Poll 5: Results

Okay, great. So the majority - the most common answer was start higher dose valganciclovir treatment. The next 1 would be start maribavir at about a third of people, and then a little bit would switch to intravenous ganciclovir or wait for resistance testing. So I think I wouldn't wait for resistance testing because it's already 18,000 going in the wrong direction - that's a pretty significant DNAemia. So going in the wrong direction.

I personally would not switch to high dose valganciclovir because it's not going to work and it almost never works. It just - there's usually a resistance mutation that's a high level resistance mutation. And almost undoubtedly things will get worse while you're waiting for resistance testing. So I actually would switch to either maribavir or foscarnet.

[00:57:25]

Management for Suspected Drug-Resistant CMV Infection 

Okay. So here's the algorithm that we mentioned from the guidelines. I will say this is the fourth version of the guidelines. Every time we publish this, this is the most popular figure in the guidelines. So there it is.

So 1, you identify that there are risk factors and that the patient likely has refractory/resistant disease. Two, you optimize the immunosuppression. You review the drug dosing. You order the resistance testing. If it's a high viral load, then you may wish to use foscarnet or maribavir. And I actually think 18,000 is a pretty reasonable—getting higher viral load. So I probably would go down the foscarnet or maribavir arm.

If it's not high, then you could do intravenous ganciclovir, valganciclovir at higher dose or maribavir.

And then important to realize that you don't want to use maribavir if there's CNS disease or retinitis. So you always want to screen for that. Those are pretty rare, but nonetheless maribavir does not penetrate into those spaces.

And then foscarnet in the guidelines. Foscarnet is suggested for high viral loads, usually greater than 50,000. Maribavir tends to work better and have lower—we believe, probably lower rates of resistance developing if there's a lower viral load to start.

And then once you have the genotypic testing, you sort of go through the algorithm and depending on the results you go through, and you pick the best therapeutic options. And depending on how the information you get looks, sometimes I just get the resistance mutation. So I have to actually find in the document - in the guidelines document, I actually find the mutation and look at how resistant it is to various antivirals etc. So you just kind of walk through the algorithm.

[00:59:06]

Patient Case #2 (cont’d)

Okay. So there she is 18,000. And this is clearly breakthrough undoubtedly pretty resistant to valganciclovir. So I actually gave her—we couldn't get maribavir quickly for her. So we actually admitted her to hospital, put her on foscarnet induction. She was found to have the UL97 mutation, the M460V, which is 1 of the top common canonical mutations.

We also switched the belatacept to ciclosporin. And as you can see, she has a good response to the foscarnet induction. And in the guidelines meeting, we actually talked a lot about maybe induction with foscarnet followed by consolidation with maribavir, not prophylaxis, but sort of starting, just really getting that DNAemia down and then consolidating with maribavir. And because there can be delays in access due to insurance, although the company has a very good program, but still sometimes there can be delays, sometimes foscarnet can give you the initial treatment option.

I personally always keep people in hospital for foscarnet, and I never send anyone home on foscarnet. And I actually have very good outcomes when it's very carefully used.

I then switched the foscarnet to maribavir and famciclovir remembering that maribavir just covers CMV. It doesn't cover HSV, VZV. And so I always think of using it as a combination pill. If somebody has resistant/refractory can be they could be at risk for disseminated zoster.

And she went home and did very well. You can see she had a very low DNAemia. And eventually after a full course of treatment, we switched her to letermovir for a full 6 months of, at this point, secondary prophylaxis.

And after we stopped the letermovir, she had a little blip of CMV at like 816 IU per ml. And we were really worried. That's very low. But we were worried given the overall context. And this is a very complicated patient who had not tolerated her immunosuppression and had a lot of things go on. So the team was quite worried.

I didn't think we needed to restart maribavir, but we did give a 1-time dose of CMV immunoglobulin. And then she subsequently resolved her CMV DNAemia and sailed off into the sunset and has done quite well.

So overall, you know, a case where we've used multiple different agents. We used the resistance testing. And the strongest risk factor for her, I think, was the belatacept treatment. And so really I can almost like smell the belatacept up in the epic chart or the EMR when I get this complex case, I'm like this patient's on belatacept. So, you know, always look for it. Sometimes it's not so obvious or not highlighted.

So I think, where possible, switch to another immunosuppressive agent while they're on the CMV treatment.

[01:01:50]

Gene Mutations in CMV: UL97 M460V/I      

Here's an example of the guidelines. This is probably hard for you to see from the back, but I just wanted to show you. So I got the information that it was the M460V mutation. And I found it in the chart. So that's a very high level ganciclovir resistance mutation. It results in an EC50 of 5- to 15-fold higher. So I'm never going to do that. I'm never going to hit that with valganciclovir/intravenous ganciclovir. So I really do need a different agent.

And then 1 of the cool things that I found out during the guidelines meeting is that this M460V or I mutation actually makes people exquisitely sensitive to maribavir. And overall, in the SOLSTICE trial, they had a statistically better response rate to maribavir compared to other ganciclovir resistance mutations. So that's kind of cool. One little good thing, right, out of all of this.

[01:02:44]

Patient Case #3

Okay. And now I'll let Atul present his case.

Dr Humar: Okay. So this is a lung transplant patient D+/R- 10 months post lung transplant. So same situation. He's on valganciclovir prophylaxis at the time this happens, but he's frequently an on and off valganciclovir because of leukopenia. Has a 2-week history of fever, fatigue, malaise.

And that's his bloodwork. Creatinine is mildly elevated at 164 or 1.85, and white count is 2.9. His CMV viral load is 35,500 IU per mL.

[01:03:22]

Poll 6

So what treatment would you switch to at this point? So:

1. IV ganciclovir;
2. Higher dose oral valganciclovir;
3. Foscarnet;
4. Maribavir; or
5. Other.

Let's see the answer.

[01:03:46]

Poll 6: Results 

Okay. So IV ganciclovir. So the majority here suggested the foscarnet and the second was IV ganciclovir.

[01:03:56]

Patient Case #3: After 2 Wk of Full Dose IV Ganciclovir 

So we actually switched this patient initially to IV ganciclovir. And as you can see, what happened is there was no clinical improvement. And that's kind of the question in the first case I presented as well. So if a patient - and the case Camille presented. If a patient's on valganciclovir and taking it at a prophylactic dose and they break through with a high viral load, a lot of those patients have true ganciclovir resistance. So, you know, if they present unwell with a high viral load, you have to be careful about using IV ganciclovir in that setting and you should consider an alternative.

But in this patient we did use IV ganciclovir and you can see the viral load went up to 62,000. And the patient did indeed have a very common CMV ganciclovir resistance mutation L595S, which also confers high level ganciclovir resistance.

[01:04:53]

Poll 7

So at this point, what would you switch the patient to? Would you:

1. Increase the dose of ganciclovir, so double dose;
2. Give foscarnet;
3. Give maribavir; or
4. Do other.

Okay. Let's see the answer.

[01:05:17]

Poll 7: Results 

Yeah. So the majority suggested foscarnet, and we would typically use foscarnet in that setting as well for that high of a viral load and the patient is clinically unwell.

[01:05:30]

Patient Case #3 (cont’d)

Although, you know, the renal function is abnormal. So you do need to be quite cautious. This patient indeed was started on foscarnet and the viral load did decrease to undetectable. But what happened is the renal function worsened.

And so the foscarnet had to be prematurely stopped. And then shortly after that, the viral load rebounded, but this time to approximately 10,000 IU per mL.

[01:05:57]

Poll 8

So what do you do now?

1. Restart foscarnet;
2. High dose ganciclovir;
3. Maribavir;
4. Letermovir; or
5. Other.

Okay, let's see the answer.

[01:06:19]

Poll 8: Results 

Yeah. So the majority here chose maribavir. And I think this is a really good setting to use maribavir. So there's a ganciclovir resistance mutation. The viral load is not terribly high. And the patient also has renal dysfunction there. So all kind of compelling reasons to use maribavir in this setting. And that's what we did.

We also decided to give the patients CMV immune globulin for 3 doses as well. And the patient had a nice viral load response. And then for secondary prophylaxis, this patient was placed on letermovir for 3 months and did well.

[01:06:55]

Summary 

So, Camille, why don't you join me up here and I'll leave the question and answer. So in summary, we hope you enjoyed this. I think it's a pretty exciting time to be in the CMV field. We're entering a new era, as you can see, where multiple diagnostic therapeutics and preventive strategies are being deployed. And this is the troll of transplantation. This is a troll that's hiding out under a bridge, and that CMV is traditionally called the troll of transplantation. And I think we're making good progress.

This is a picture of the guidelines group that met in Montreal last June. So thank you very much. And we'll do some questions.

[01:07:34]

Q&A

Dr Kotton: Great. So 1 of the first questions I think is a pretty straightforward one, but is if CMV viremia worsens after starting treatment, from what time point should you count 2 weeks to determine refractory CMV? Is it the baseline or is it the peak value?

And I usually say it's from the time that we initiated treatment. Okay. So it's 2 weeks from the initiation of treatment. The peak is a migratory thing. So you don't know when the peak is are going to happen. So it's sort of 2 weeks from the onset of treatment.

Dr Humar: Do you want us to do the post-test questions first?

Dr Kotton: Oh okay. Sorry.

Dr Humar: Yeah. Why don't you do that?

[01:08:10]

Posttest 2

Dr Kotton: Sorry, sorry. So a heart transplant - heart transplant patient has breakthrough CMV on appropriately renally adjusted valganciclovir prophylaxis. She's compliant. The viral load is 110,000 on plasma. So what is your next therapeutic choice? So this will be interesting to see if you've changed your answers. Would you:

1. Double the dose of valganciclovir;
2. Start intravenous ganciclovir;
3. Foscarnet;
4. Letermovir; or
5. Maribavir.

Go ahead and vote. Go ahead and show us the answers.

[01:08:53]

Posttest 2: Results 

Okay, so it looks like a lot more people are starting foscarnet. So guidelines, we said, you know, above 50,000 IU per ml probably start foscarnet. Probably, you know, less maribavir in that setting. I'm glad nobody's using letermovir for treatment. So that is - that is great.

[01:09:17]

Posttest 2: Rationale 

And doubling the dose of valganciclovir is probably not going to be effective. And it is a high viral load. So we're worried this patient could be getting sicker. So great. Excellent. Lots of learning there.

Okay, so now the Q&A.

[01:09:32]

Q&A

Dr Humar: Okay, so the next question. Yeah. So when you're treating patients with CMV, I think we both said you could treat either to negative or you could aim for a center specific threshold. So what - what does your center, Camille, consider as that threshold to aim for if it's not zero?

Dr Kotton: Yeah. So you know, I often think - so we test on plasma. One of the big issues is if you're using a whole blood assay. And it can actually be hard to tell. Like you have to look pretty hard sometimes in the computer.

Whole blood tends to give a higher overall result. So with whole blood, you may not get to negative. So you may treat for - you could treat for months, but you may not achieve negative. It is a latent virus. It's still in the body. You may still detect it. So that's an important branch point for me, is if it's whole blood, I’ll look at the higher number than if it's plasma. But usually when it's below 200, I'm a lot less worried. I'm a lot less worried about CMV.

So I will often take that as if it's double digits or below 200, I'm often like, yeah, we've treated well enough. The body will, you know, mop up the rest. I think that many people keep treating for weeks trying to get the 152 down to, you know, 78, like that's not really necessary. In the guidelines, we talk a lot about what is a clinically significant change. And we recommend thinking about it in log.

And then that it should be a 0.7 log base 10 change when it's especially when the it's below 1000. So a lot of those numbers are sort of like the same number actually using real-time PCR. And so I think don't over treat. I think we do end up over treating. And then the patients it's expensive and then they can get the cytokines especially leukopenia from the valganciclovir.

What do you think? I'd love talking to you about this question.

Dr Humar: Yeah, I think this is 1 of the most common traps that people fall into, is trying to get the viral load down to zero. And it's just not possible in many patients. And so at our own institution, we use a threshold of 200 IU per ml in plasma. And anything below that, we just call negative actually. And we actually ask the lab to report those as, as less than 200 and not quantify those.

And I think if you do something like that, if you can set a center specific threshold, it will help you immensely in interpretation of these ultralow viral loads. You know, people are trying to treat 50 IU per mL, 35 IU per ml, and they're wondering if they should start maribavir at 75 IU per mL and things like that.

And what happens if with these grumbling little low level viremias is many patients will then develop immunity. They'll develop a cellular immune response over time, and they'll actually be able to eliminate that low level of viral load. Although not all patients, you know, some will obviously recur, but many patients will.

Dr Kotton: Often when I have those low level results, I may stop therapy, but I might do some surveillance after treatment. So then I start doing—you know, especially if it's somebody that I think is high risk for a variety of factors. I'll just start doing weekly monitoring and make sure I don't miss the opportunity to treat them at a low level, but I tend to be shorter. I actually think I tend to be shorter on treatment, and I love what he said about his laboratory and how he asked them to report less than 200 as negative.

I asked my laboratory if they'd do that and they said no. Life is better in Canada, I guess.

Dr Humar: This is a common question that gets asked is checking CMV and BAL fluid in the ICU patients?

Dr Kotton: Yeah. So we covered this in the CMV definitions. And actually, it turns out I was the one in that definitions group that surveyed all these lung transplant programs, including his program in Toronto and my program at Mass General and Duke and multiple other big lung transplant programs.

And it turns out that CMV PCR on BAL is generally really not useful. The higher the number, the more likely they have pneumonitis. But honestly, it depends. BAL is by definition lavage, right? So it depends on how much fluid is put in. We get a lot of numbers, you know, 10 to the 6, 10 to the third, all kinds of things. But if it's really high, you can diagnose pneumonitis that way. But otherwise it's really not useful.

And what I found out in my survey is that most of these big transplant programs don't test. Many - so that is actually really helpful. So we largely have stopped doing - I've stopped recommending doing BAL CMV PCRs. That's really - it's a clinical diagnosis. You know it when you see it. You know, the patient has pneumonitis. They often have a significant CMV DNAemia, you know, but also people will shed CMV. And so you don't want to treat all these people that are just shedding because they are in the intensive care unit. So in general we're sort of refraining from sending that.

Dr Humar: Yeah. Yeah, we were doing it and eventually we had to ask the lab to stop doing it, because we just didn't know how to interpret all these positive results that we were getting. So I think it's better to just rely on the peripheral viral load and the clinical picture. In the vast majority of settings, there might be some extenuating clinical circumstances where you may specifically want that test.

[END OF TRANSCRIPT]