2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Sameer A. Parikh, MD, discusses how the International Prognostic Index has assisted physicians with their patients who have CLL, the optimal sequencing of therapies, and other emerging molecular targets in this disease.
Sameer A. Parikh, MD
The therapeutic armamentarium of chronic lymphocytic leukemia (CLL) has quickly expanded over the years beyond the standard fludarabine, cyclophosphamide, and rituximab (FCR) regimen to include promising targeted options, such as ibrutinib (Imbruvica), idelalisib (Zydelig), and venetoclax (Venclexta).
How physicians can decide on an agent and when to administer it can be based on the CLL International Prognostic Index (IPI), a scoring system that tests for the common prognostic factors of the disease. Results of a study published in The Lancet Oncology demonstrated that the CLL-IPI allows for more targeted management of patients with CLL in both clinical practice and in trials exploring novel agents.
“For all your patients with CLL, try to obtain the CLL-IPI because that is now becoming an effective tool for our ability to prognosticate these patients and predict when they are going to get treatment,” said Sameer A. Parikh, MD.
OncLive: Please provide an overview of your lecture on CLL.
Parikh, an assistant professor of medicine at Mayo Clinic, discussed how the IPI has assisted physicians with their patients who have CLL, the optimal sequencing of therapies, and other emerging molecular targets in an interview during the 2017 OncLive® State of the Science Summit on Hematologic Malignancies.Parikh: I broke down the talk into several different parts. The first part was prognosis in patients with CLL who are newly diagnosed and don’t yet meet indication for therapy. Roughly two-thirds of all our patients at the time of diagnosis are early stage, asymptomatic, and do not need any treatment. There are a myriad of prognostic markers available in this particular field to be able to tell our patients when they are likely to need treatment.
How do you choose a frontline therapy for your patients?
One of the most important contributions to the literature has been the development of the so- called IPI, or the CLL International Prognostic Index. This combines 5 different factors, including age, clinical stage, TP53 mutation status, IGVH mutation status, and serum β2-microglobulin concentration. This will allow us to prognosticate when these patients are going to need therapy. We try do clearly obtain this for all of our patients, [especially for those who are] newly diagnosed and don’t need treatment yet. Frontline therapy in CLL has evolved over the years and, with the introduction of novel targeted therapies, this has changed quite a bit. One of the most important [approaches] that has come out is targeting the BCR pathway, [with treatments such as] ibrutinib, idelalisib, and BCL-2 antagonists, including venetoclax. There are several major phase III clinical trials that have compared conventional cytotoxic chemotherapy—including FCR and bendamustine and rituximab (Rituxan; BR)—to novel agents. The results of these are all pending.
We tried to individualize treatment for our patients based on what their risk stratification is, and [if there] is evidence of 17p deletion—then, cytotoxic chemoimmunotherapy is not really considered a good option for these patients. We try to move to some of the other novel therapies, such as ibrutinib, because there are excellent data for that medication in those groups of patients.
If the patient is young with mutated IGVH status, there are now data that indicate long-term remissions from FCR therapy and so we try to use a particular combination in those select groups of patients. The clear majority of our patients are older, have a number of comorbidities and, in that area, there are data to suggest that combination therapy of chlorambucil and obinutuzumab (Gazyva) leads to excellent progression-free survival (PFS) and overall survival compared with chlorambucil alone and chlorambucil and rituximab. That is a very good treatment option.
With the emergence of newer agents being developed and moving along the pipeline, are there any concerns with sequencing?
The most recent data in the upfront setting was the RESONATE-2 trial, which compared chlorambucil with ibrutinib and showed that ibrutinib had a very good PFS. The field of treatment has expanded considerably and now, thankfully, we have many treatment options to offer our patients. That is an excellent question. Sequencing is a very important question; we all have seen data that patients who received cytotoxic chemoimmunotherapy with FCR or BR evolved and developed more resistance in their disease process, with acquisition of other genetic markers that are harder to treat. Therefore, sequencing is a very important thing in the planning of CLL therapy for our patients.
The optimal sequence has obviously not been determined yet; we are lagging behind our colleagues in multiple myeloma, where they have looked at the sequencing. We, hopefully, are developing clinical trials where we will determine what the optimal sequence is. Among patients who have 17p deletion, there is no question that those patients should be started off with novel agents because traditional cytotoxic chemoimmunotherapy does not work for them.
Aside from 17p deletion and TP53 mutations, what are targets are being explored?
We do not know if patients without 17p deletion who were started on ibrutinib would have a response to FCR or BR at the time of progression. This is because we just don’t have that long amount of follow-up in these patients to be able to confidently state that. Sequencing is certainly an important aspect of management but, right now, there is very little data to guide how best to determine the exact sequence of treatment. Over and above the CLL-IPI that takes into account 17p deletion and TP53 mutation, there are data from whole-exome sequencing and other groups that have looked at only genetic markers to be able to predict prognosis in patients with CLL. There are many recurrent mutations that have been described, including NOTCH1 and SF3B1, to name a few. All of those mutations are prognostic, meaning they will tell us whether a patient is going to do poorly or not, and they all segregate within specific subgroups of cytogenetic abnormalities.
For example, NOTCH1 is mostly seen in patients with trisomy 12, and SF3B1 is preferentially seen in patients with 13q deletion. Each of these mutations we know is deleterious and is associated with poor prognosis. The challenge is to be able to make them into a predictive biomarker, meaning I know that they are going to affect the prognosis in a negative way, but I might be able to do something to fix that.
Could there be room for checkpoint inhibitors as a treatment for patients with CLL?
That is a predictive aspect of things, which we don’t have any medications for. The TP53 mutation is both prognostic and predictive, meaning I know that they are associated with poor outcomes and, at the same time, I know that if I use traditional FCR, those patients are not going to do well. I need to use ibrutinib in patients who have the TP53 mutation. We don’t have similar drugs for NOTCH1 or SF3B1, which are all associated with poor prognosis—but we need to develop those drugs. Our group published the first trial of PD-1 inhibitor therapy in CLL and in Richter's transformation. This was led by Dr Wei Ding and this was a phase II study of single-agent pembrolizumab (Keytruda) given every 3 weeks at standard doses. Of the 25 patients who were treated on this trial, 16 had relapsed/refractory CLL and 9 had Richter’s transformation of their CLL. Unfortunately, there were no responses seen in patients with CLL alone.
On the other hand, there were impressive responses seen in patients with Richter’s transformation. The response rate was about 44%, which is pretty good in a cohort of patients who had transformed after being on ibrutinib, where, historically, the outcomes have been very poor. The median survival in these patients was more than 11 months, where, historically, it has been less than 3 months. The duration of responses has also been equally good.
What are the key takeaways from this presentation?
There is a larger trial now being planned that is an expansion cohort currently accruing patients for Richter’s transformation at Mayo Clinic. We are hopefully going to accrue several additional patients to the Richter’s transformation arm, and there are many correlative studies that are being planned to determine why the patients with CLL don’t respond but the Richter’s transformation patients do. All of that work is still ongoing. In your young, fit patients who are candidates for FCR therapy, you should strongly consider doing that, particularly if they have mutated immunoglobulin genes. This is because the remission duration in many studies now exceeds 15 years.
This contrasts with a daily therapy like ibrutinib or other novel agents that need to be taken on a continuous basis. In patients who are elderly and have several comorbidities, although there are no head-to-head comparison trials, either a combination of chlorambucil and obinutuzumab or ibrutinib alone are perfectly appropriate choices. Before you treat anybody for CLL with any particular therapy, consider checking their TP53 mutation status, because it has important treatment implications.
The International CLL-IPI working group. An international prognostic index for patients with chronic lymphocytic leukaemia (CLL-IPI): a meta-analysis of individual patient data. Lancet Oncol. 2016;17(6):779-790. doi:10.1016/S1470-2045(16)30029-8.
Related Content: