Tumor Lysis Syndrome in Patients With Chronic Lymphocytic Leukemia

Josie Montegaard

Amber, why don’t you lead off our conversation today by giving some background on tumor lysis syndrome in the CLL setting and which patients are at highest risk. Why is it so important to stay ahead of it?

Amber Koehler

Sure, I’m happy to. Tumor lysis syndrome is a possible adverse event that can be seen in patients with CLL, particularly those undergoing treatment with venetoclax-based regimens. TLS is characterized by rapid breakdown of CLL cells; as these cells die off, they release potassium, phosphorus, and nucleic acids, which are converted into uric acid, into the bloodstream. The phosphorus binds calcium, which can result in hypocalcemia. Hyperuricemia can cause renal injury, as well as other issues. If a patient is hyperkalemic, they can have cardiac arrhythmias; if they are hypocalcemic, they can have muscle spasms. TLS can occur in either the frontline or the relapsed setting.      

Venetoclax was originally approved in the relapsed setting, and it was typically started concurrently with rituximab therapy. Because of this, there was often a high disease burden as we brought venetoclax on board. The initial observations of TLS in patients with CLL in some of the early venetoclax-based trials changed how we give venetoclax—now, we do a weekly dose escalation over the course of 5 weeks to get patients up to the full dose of 400 mg daily.¹ The dose increases once weekly with close monitoring for TLS, starting with 20 mg daily for 1 week, 50 mg daily for 1 week, 100 mg daily for 1 week, 200 mg daily for 1 week, and finally 400 mg daily as the ongoing dose. Administration of venetoclax in this manner really helps mitigate the risk for TLS.

Josie Montegaard

As you said, the advent of the dose escalation schedule has helped decrease the risk of TLS. Another way we can help prevent it is by stratifying patients’ risk. How do you all approach risk stratification for TLS?

Amber Koehler

One way to approach this question is by using the package insert for venetoclax. Risk is defined as low, medium, or high, based primarily on the absolute lymphocyte count and the size of lymph nodes. Patients who have no lymph nodes greater than 5 cm and an absolute lymphocyte count of less than 25,000 are considered low risk. Patients with lymph nodes ranging from 5 to 10 cm or an absolute lymphocyte count greater than 25,000 are considered medium risk. Patients with a conglomerate lymph node mass, a single lymph node that's greater than 10 cm, or an absolute lymphocyte count greater than 25,000 in addition to any lymph nodes that are greater than 5 cm are considered high risk.

The National Comprehensive Cancer Network Guidelines² and International Workshop on Chronic Lymphocytic Leukemia criteria,³ most recently updated in 2018, say that computed tomography (CT) scans are optional in terms of restaging. However, I would argue that it is important to have patients get a CT scan, especially if you're going to start them on venetoclax-based therapy. Although we can often palpate lymph nodes in certain regions such as the neck, armpits, or groin, there is the possibility of large conglomerate mesenteric adenopathy or retroperitoneal nodes that could be missed without a CT scan.

It's important to have a good understanding of a patient’s disease bulk in order to accurately risk-stratify them. Massive splenomegaly can be a potential contributing factor for TLS.⁴ They don't include that in the full risk stratification criteria, but it is included as a side note in the venetoclax package insert. Because of this, providers might consider upping the TLS risk category for patients with massive splenomegaly from CLL.

Josie Montegaard

I completely agree with repeating the CT scans prior to starting the venetoclax ramp up. As you said, big lymph nodes that we’re never going to feel can be hidden in the abdomen. With the up-front use of venetoclax in combination with obinutuzumab, it's important to get those CT scans.

Timing with the CT scans is also very important. We usually schedule the CT scan after patients have gotten the first few doses of obinutuzumab, right before the week that they're going to start venetoclax. Even those first few doses of obinutuzumab can really debulk patients’ disease burden and bring them down from high risk to a medium- or low-risk category. Therefore, it’s important to put some forethought into the entire plan so that we don’t order the CT scan right before beginning the obinutuzumab and then need to get insurance approval for another CT scan 3 weeks later.

Amber Koehler

Absolutely. A lot of the newer regimens in clinical trials are looking at lead-ins with drugs like Bruton tyrosine kinase (BTK) inhibitors. The idea is that if we lead in either with BTK inhibition (with ibrutinib or acalabrutinib, etc.) or obinutuzumab as used in the CLL14 regimen,⁵ it can help cut down on the risk for TLS and hopefully prevent some hospital admissions.

Josie Montegaard

Another step we take to minimize the risk of TLS when we're preparing someone to begin a venetoclax-based treatment is we will start them on an antihyperuricemic agent, such as allopurinol, usually at least 3 days (but a week or more if we can) before they begin the venetoclax.

Additionally, we'll have them increase their oral hydration to at least 2 to 3 liters a day. I always tell my patients that it's okay to front-load the fluids during the day if they’re worried about being up all night going to the bathroom. If that's a real barrier, having patients push the fluids until 5:00 PM and then just taking sips of water thereafter can be greatly helpful.

It is also important to make sure that we understand what patients’ actual TLS risk is so that we can prepare the appropriate escalation setting. At Dana-Farber, for patients who are considered low and medium risk, we typically ramp up treatment in the outpatient setting. Sometimes we will admit medium-risk patients for the first and second dose escalations if they have a risk factor that makes us worry about TLS. For example, if the patient has poor renal function and venetoclax is the best treatment option for them, we may consider an admission just to see how it goes. Admitting patients and monitoring their labs may be burdensome for the patient, but it is better than the risk of missing TLS in the outpatient setting.

For our high-risk patients, we'll admit them to the hospital, at least for the first and second dose escalations. If they've done okay, sometimes we'll consider the remainder of the escalations to be outpatient. Is that similar to what you all do at Mayo?

Amber Koehler

I agree. We recommend baseline TLS labs prior to each dose escalation, as well as TLS labs 6 to 8 hours after the first dose at the increased dose level and again 24 hours after that first increased dose. The highest risk for TLS appears to be at the 20 mg and 50 mg dose escalation. Usually, once they get up to 100 mg dose escalation and beyond, the 24-hour TLS labs after that first higher dose typically suffice. However, there are exceptions, so it is important to look at the individual patient to make that decision based on their risk stratification.

The logistics of this can get a little tricky depending on the turnaround time of the lab. Let's say that the patient comes in, gets labs, sees their provider in the morning, and takes their medication in the office. They now need repeat labs 6 to 8 hours later. Depending on how late in the morning their appointment was, this might be a bit of a challenge both in terms of turnaround time for the labs and also in terms of who will be responsible for acting on those lab results if they come back after routine clinic hours. These logistical challenges are one reason that patients may need to be admitted to the hospital for TLS monitoring with dose escalation of venetoclax. Certain practices have a hospital-based outpatient setting, which we’ve used for some of these patients as well.

Josie Montegaard

When we are monitoring the labs, as you mentioned, our standard practice is to get a full set of chemistries before dosing in the morning to establish a baseline. This includes a comprehensive metabolic panel, uric acid level, lactate dehydrogenase (LDH), and a phosphorous level. We’ll repeat those labs 6 to 8 hours after they’ve dosed, and then again 24 hours after they have dosed.

I want to highlight that to meet the laboratory definition of TLS, a patient does not need to have crazy elevations in electrolytes. When we hear TLS and see the data on how TLS occurred in the early venetoclax trials, we might expect to see a potassium level go from 4 to 6.5 or an LDH to go sky high.

In actuality, TLS in the outpatient setting—or at least laboratory TLS—can be quite subtle. It's detecting those subtle changes that allows us to act quickly and make sure that laboratory TLS does not escalate to clinical TLS.

The Cairo Bishop classification for TLS⁶ does cite quite high elevations in uric acid and potassium, but they also include the caveat of a 25% increase in the patient’s baseline. Therefore, if someone's baseline potassium is 3.5 and now all of a sudden they're at 4.5 and other electrolyte levels are similarly elevated, they could be undergoing TLS despite the fact that they're still within the range of what is considered normal for that level. If we saw an elevation like that and felt that laboratory TLS was occurring, we would start a fluid bolus—we’d give intravenous or oral furosemide, depending on how high the potassium level gets.

We also can give patients polystyrene sulfonate for hyperkalemia or rasburicase in the clinic for an elevated uric acid. We are starting to intervene sometimes even if electrolyte levels are still normal. It's all about being proactive in these instances to keep these patients from actually developing organ dysfunction because of TLS.

Amber Koehler

You bring up a great point about the difference between laboratory and clinical TLS. It is important to recognize that it's more than just the absolute value—it's also the trend.

Because we want to avoid giving rasburicase to patients with G6PD deficiency, do you all, as part of your initial workup or restaging, check the G6PD level in all high-risk patients before beginning them on venetoclax?

Josie Montegaard

No, we typically don't. To be honest, even with our high-risk patients, because we are starting them on their oral antihyperuricemic agent quite early on and we will admit them the night before a venetoclax escalation, hydrate them through the night, and then check labs every 4 hours, I haven’t found that rasburicase has been needed all that often. We actually use it more in the outpatient setting, when, for example, a patient who is medium risk for TLS surprises us and does start showing some uric acid elevations. But usually at that point, we haven’t tested G6PD levels—it's a one off.

Amber Koehler

That has been our experience as well. It's not something we commonly test for because we don’t often need to use rasburicase.

Josie Montegaard

Another important point is obtaining an accurate baseline set of chemistries. Something that plagues me quite often is pseudohyperkalemia, which is when potassium levels can be falsely elevated due to a very high white blood cell count, which increases the risk for homolysis during the blood draws.

Pseudohyperkalemia is something to be mindful of because we need to get an accurate potassium level at baseline to understand if someone is developing TLS at their 6-hour labs. Similarly, if someone has their 6-hour labs drawn and their potassium level is very high, it could be due to pseudohyperkalemia. It's important to recognize that because we don't want to treat a pseudohyperkalemia when it's actually a normal potassium level and thus cause the patient to actually develop a hypokalemia.

To try to prevent this, we always do a regular blood draw initially. If the labs do come back suggestive that the patient has pseudohyperkalemia or are noted to be hemolyzed, we will redraw the lab. When we redraw, we'll do it as a separate, fresh blood draw, and not off a preexisting IV. We'll try to avoid using a tourniquet, and then we’ll bring the sample on ice directly to our laboratory for processing to try to prevent the hemolysis.

Additionally, the use of arterial blood gas tubes or whole blood potassium tubes can be helpful to minimize the risk of hemolysis. Those are some tricks of the trade that we use to help minimize this problem, or at least to help obtain an accurate level if it does occur. Do you all do something similar?

Amber Koehler

Yes, that has been our experience as well. We see it most frequently with those first couple of doses, and then it becomes a non-issue.

You referred to a few of the ways we treat hyperkalemia. It’s important to remember that the different interventions we use to treat hyperkalemia have varying onsets of action and durations. Ideally we’d choose an intervention that is quick acting in combination with something that has a more durable response.

One topic we haven’t covered yet is how we manage hyperphosphatemia. In general, I don't see many symptoms related to the hyperphosphatemia. If anything, it can compound the renal injury. Usually we can treat it with calcium acetate (PhosLo) and by educating patients on eating a low-phosphorus diet. Has that been your experience as well?

Josie Montegaard

We typically start patients on an oral phosphorus-lowering agent. We love hydration here, so any chance we get to give patients extra hydration, we will. We’ll also instruct patients to modify their diet, as you mentioned.

Amber Koehler

As long as we can correct the high phosphorus, then the hypocalcemia is not generally something that we need to intervene on. The exception would be if patients are becoming symptomatic from it. I haven't had any patients personally who have had really severe hypocalcemia where we would need to think about repletion. What have you seen in your practice?

Josie Montegaard

I agree. I think most commonly we're correcting the other electrolyte arrangements rather than a low calcium level.

Amber Koehler

One point that I think is important for us to discuss is the frequency with which TLS occurs.

The package insert states that it can occur in approximately 2% to 5% of patients—these values include both laboratory and clinical TLS. Real studies have suggested similar rates, ranging from around 3% to 6% to up to 13%.^4,7

Many people are familiar with Cairo Bishop criteria for TLS.⁶ However, the studies that looked at rates of TLS used the Howard criteria,⁸ which is even harder to qualify for (for both for laboratory and clinical TLS). The Cairo Bishop definition of laboratory TLS requires two or more abnormalities, whether it's the absolute value or the 25% increase within a period of 3 days prior to 7 days after the dose escalation. The Howard criteria has the same requirement, but those two abnormal values must be within 24 hours of each other.

Even though the risk for TLS might seem intimidating, thankfully the mitigation strategies we have available to us including appropriate risk stratification and intervention, the weekly dose escalation schedule, and some regimens that lead in with other agents to debulk disease prior to initiation of venetoclax make this risk pretty manageable.

Josie Montegaard

As you said, the advent of the dose escalation schedule alone has been helpful. With all these newer trials coming out, having the lead-in treatments prior to venetoclax means that by the time patients get to venetoclax, in many cases their blood counts are normal, and they hardly have disease on their CT scan. So the risk really is quite minimal.

We still follow all of the precautions because we’ve had surprise instances where patients end up having some tumor lysis. The key is knowing what to look out for, taking preventative measures, and hydrating patients as much as possible throughout.