CME
Physicians: Maximum of 1.25 AMA PRA Category 1 Credits™
Released: May 10, 2022
Expiration: May 09, 2023
John M. Burke, MD:
The phase III POLARIX trial evaluated the safety and efficacy of polatuzumab vedotin with chemotherapy in previously untreated DLBCL.1,2 Polatuzumab vedotin is an antibody–drug conjugate consisting of an anti‑CD79b antibody linked to the microtubule inhibitor monomethyl auristatin E. Polatuzumab vedotin is approved for use with bendamustine and rituximab (R) in patients with R/R DLBCL.3
A previous phase II trial combined polatuzumab vedotin with cyclophosphamide/doxorubicin/vincristine (CHP), leaving vincristine out of the typical cyclophosphamide/doxorubicin/vincristine/prednisone (CHOP) regimen to minimize overlapping toxicity of peripheral neuropathy, which can occur with both vincristine and polatuzumab vedotin. In that phase II study, polatuzumab plus CHP in combination with either R or obinutuzumab showed activity and a manageable safety profile in previously untreated DLBCL.4
In the double-blind, placebo-controlled phase III POLARIX trial, patients with previously untreated DLBCL were randomized to receive six 21-day cycles of polatuzumab vedotin plus R‑CHP plus a vincristine placebo or R-CHOP plus a polatuzumab vedotin placebo. Polatuzumab vedotin was given at a dose of 1.8 mg/kg IV with each cycle. All patients received an additional 2 cycles of R. The trial enrolled 879 patients, and the primary endpoint was PFS as assessed by the investigators. Secondary endpoints included event-free survival (EFS), CR rate at the end of treatment, disease-free survival (DFS), overall survival (OS), and safety.
John M. Burke, MD:
The 2 treatment arms were well balanced. The median age of the patients was 65‑66 years, and the median time from diagnosis to treatment was 26‑27 days; 88% to 89% of patients had Ann Arbor stage III or IV disease; 62% had an International Prognostic Index (IPI) score between 3 and 5; and approximately one third had activated B-cell subtype of DLBCL.
John M. Burke, MD:
The overall response rates were not significantly different between the 2 treatment arms at 85.5% with polatuzumab vedotin plus R‑CHP vs 83.8% with R-CHOP. The CR rates were 78% in the polatuzumab vedotin arm and 74% in the R-CHOP arm, whereas the partial response rates were 7.5% and 9.8%, respectively.
John M. Burke, MD:
The 24‑month PFS was 76.7% in the polatuzumab vedotin plus R‑CHP arm vs 70.2% in the R‑CHOP arm. This had an HR of 0.73 (95% CI: 0.57-0.95; P = .02), indicating a 27% reduction in the risk of progression or death with polatuzumab vedotin plus R‑CHP.
EFS—which was defined as disease progression or relapse, death, administration of new anti‑lymphoma treatment that was not specified in the protocol, or biopsy‑confirmed residual disease after the end of treatment—was similarly improved with polatuzumab vedotin plus R‑CHP vs R-CHOP, with an HR of 0.75 (95% CI: 0.58-0.96; P = .02).
John M. Burke, MD:
DFS was also improved with polatuzumab vedotin plus R‑CHP compared with R-CHOP, with an HR of 0.70 (95% CI: 0.50-0.98). However, OS was not different between the 2 treatment groups, with an HR of 0.94 (95% CI: 0.65-1.37; P = .75).
John M. Burke, MD:
More patients in the R‑CHOP arm needed radiotherapy, stem cell transplant, and CAR T‑cell therapy as salvage treatments. Overall, fewer patients on the polatuzumab vedotin plus R‑CHP arm needed subsequent antilymphoma therapy than those who received R-CHOP, with 22.5% and 30.3% of patients in the 2 groups, respectively, requiring additional therapy.
John M. Burke, MD:
The safety profiles of the 2 treatments were comparable with similar rates of most adverse events, including peripheral neuropathy. Polatuzumab vedotin plus R‑CHP did have a higher rate of febrile neutropenia, which occurred in 14.3% of patients (13.8% grade 3/4) compared with 8% of patients receiving R-CHOP (8% grade 3/4).
John M. Burke, MD:
Overall, the rate of grade ≥3 adverse events was similar between the 2 treatment groups. Serious adverse events occurred in a slightly higher percentage of patients receiving polatuzumab vedotin plus R‑CHP, but dose reductions were required in fewer patients in this group, although drug discontinuations were similar in both groups.
John M. Burke, MD:
An additional subgroup analysis of PFS reported in the published results of this trial demonstrated that outcomes varied between several patient subgroups. Patients with bulky disease, germinal center B-cell-like subtype, age younger than 60 years, or an IPI score of 2 did not seem to derive benefit from polatuzumab vedotin plus R CHP compared with R-CHOP, whereas those without bulky disease, with activated B-cell subtype, age older than 60 years, or an IPI score of 3 5 benefited more from the addition of polatuzumab vedotin to R CHP.
John M. Burke, MD:
To summarize, treatment with polatuzumab vedotin plus R‑CHP improved several endpoints compared with R-CHOP, including the primary endpoint of PFS and secondary endpoints of EFS and DFS. However, other endpoints, including CR rate and OS, were not improved. The need for additional therapies, including radiation, stem cell transplant, and CAR T‑cell therapy was lower in patients on the polatuzumab vedotin plus R‑CHP arm. Toxicities were similar, with patients in the polatuzumab plus R‑CHP arm showing slightly more febrile neutropenia than R-CHOP but no difference in peripheral neuropathy and less need for dose reductions.
Pending regulatory approval, in addition to entry into national guidelines, I think polatuzumab vedotin plus R‑CHP will likely become a standard option for patients with previously untreated DLBCL. However, given the lack of OS benefit and the higher cost of polatuzumab vedotin plus R-CHP, R-CHOP may still have a role for selected patients.
With regard to the additional subgroup analysis of PFS, it is really challenging to know how to apply that information to clinical practice. For example, do patients have to meet only one of those criteria, all of those criteria, or somewhere in between to derive benefit? It is important to remember that subgroup analyses like this are not powered to demonstrate differences between subgroups, so that results of such analyses are generally hypothesis generating, not definitive.
Peter Martin, MD:
Simply put, the POLARIX study met its primary endpoint. Polatuzumab vedotin plus R-CHP significantly improved PFS, which historically has not been an easy feat in DLBCL. Moreover, the addition of polatuzumab vedotin to R-CHP resulted in an improvement in efficacy without really changing the safety profile compared with R-CHOP. Of note, this has not been easy to achieve in DLBCL clinical trials. I think we all have to acknowledge that the trial was somewhat disappointing in that it did not clearly affect OS, but it was ultimately a positive trial. Put differently, for every 12-13 people treated with the polatuzumab vedotin-containing regimen compared with standard R‑CHOP, one less patient required a different subsequent therapy. From a patient’s perspective, that equates to a meaningful benefit. It will take a little more time to dive deeper into the subgroup analyses and identify which patients are most likely to benefit from polatuzumab vedotin plus R-CHP, but this regimen appears to be better than R‑CHOP, and I think it should probably be considered in the untreated DLBCL population that was studied in the POLARIX trial.
Peter Martin, MD:
Although many patients with DLBCL achieve a CR after first-line therapy, some will experience molecular relapse. There is interest in detecting relapse early to be able to improve outcomes, but it has not been clear whether any benefit of posttherapy surveillance imaging outweighs the potential drawbacks. A retrospective analysis suggested that MRD monitoring during DLBCL surveillance might identify the risk of recurrence prior to any clinical evidence of disease.5 At ASH 2021, Kumar and colleagues6 presented the results from a study that prospectively looked at the use of a next-generation immunosequencing assay for MRD surveillance in DLBCL.
The study enrolled 500 patients with acute de novo DLBCL or other high‑grade B-cell lymphomas. All patients had some baseline blood and tumor tissue collected to validate the clonotype MRD assay. Patients who achieved a PET‑negative CR rate entered a surveillance period with routine CT imaging every 6 months and blood samples collected every 3 months for MRD analyses for 2 years in the posttreatment setting. MRD positivity was defined as any detectable rearrangement, and undetectable MRD was defined as no evidence of rearrangement. The primary endpoint was characterization of the immunosequencing MRD assay, specifically the sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV).
Peter Martin, MD:
Of the 500 patients who enrolled in the study, 398 were evaluable for the surveillance phase of the study. The unevaluable patients included those who did not achieve a CR to the primary therapy or did not have pretreatment tumor pathology available. Of those 398 evaluable patients in surveillance, 43 (10.8%) of them experienced a recurrence within the first 2 years. Of interest, nearly one half (47%) of those recurrences were detected by a standard approach using clinical symptoms and surveillance imaging and/or evaluation by an oncologist, whereas for most of the other patients (44%), disease relapse was confirmed by surveillance imaging alone.
Peter Martin, MD:
Of the 43 patients with relapsed disease, 5 (11.6%) patients were not evaluable for the MRD assay; 3 due to failed quality control, 1 failed calibration, and 1 did not have MRD results within 90 days of relapse. Among the 38 (88.4%) patients who experienced disease progression and had MRD results, 19 were MRD positive prior to clinical relapse, and 14 remained MRD negative despite clinical relapse.
Peter Martin, MD:
For various reasons, including failed quality control and failed calibration, approximately 22% of patients who achieved a CR did not have evaluable MRD results. This obviously complicates the use of MRD in a true clinical setting. Out of the 356 patients in CR, 279 (78.4%) patients had evaluable MRD results, of which 78% were MRD negative throughout CR and 22% were MRD positive either in peripheral blood mononuclear cells (PBMCs) or plasma despite having a durable clinical CR.
Peter Martin, MD:
Unfortunately, with regard to the primary objective of the study, the PPV of MRD testing was quite low, although it was slightly higher in plasma compared with PBMCs. Only a minority of patients with a positive MRD result experienced clinical progression. Fortunately, the NPV was higher, and similar results were obtained in the plasma and PBMCs. Although an NPV of 94% is high, it is not perfect because approximately 6% to 7% of patients experienced disease recurrence despite having a negative MRD test.
Peter Martin, MD:
There are a few key conclusions from this study. First, the PPV of MRD testing in this study was somewhat disappointing. In addition, there were a number of technical challenges that limited the utility of MRD testing. Although the NPV of MRD testing was much higher, it did not equate to a better predictor of survival compared with simply monitoring patients for clinical symptoms and using surveillance imaging. There is still work to be done to figure out where MRD testing fits in DLBCL surveillance.
To me, one of the more interesting—and unintended—observations in this study was the role of surveillance imaging in DLBCL. Almost half of patients with relapsed disease were detected by surveillance imaging alone in the absence of clinical symptoms. I think this observation could potentially challenge the recent standard of care, which is to stop surveillance imaging in patients who achieve a PET‑negative CR at the end of therapy. On the other hand, it is worth noting that only 19 out of 398 evaluable patients (5%) of the full study population had a relapse detected by surveillance imaging alone. So, if we were to image all patients who achieved a CR, we would be imaging a lot of patients to detect a small number of recurrences. However, this number could be increased in a higher‑risk population of patients. This approach could be a way to use surveillance imaging in a higher-risk patient population and could be considered in the future. Unfortunately, this study does not provide information about whether early detection of relapse, either by imaging or MRD testing, will ultimately affect clinical outcomes. That will require a separate study.
John M. Burke, MD:
I agree that the PPV results were disappointing and that 44% of the relapses were detected by surveillance imaging alone. Many retrospective studies have not conferred much value to surveillance imaging, and the argument has been that patients often detect their own relapses, so scans are not especially useful for detecting relapse. Instead, this trial prospectively found that a significant percentage of relapses were detected by surveillance scans. This observation raises the question of whether surveillance scans may be useful in DLBCL, particularly now that CAR T‑cell therapies could become an option in second-line treatment. It is worth noting that at least one of the studies that showed significant benefit with CAR T‑cell therapy, the ZUMA‑7 trial, did not allow chemotherapy as bridging therapy and therefore excluded patients who had an urgent need for therapy.7 In theory, there could be value for earlier detection of DLBCL relapse to allow patients to move to CAR T‑cell therapy quickly without a need for bridging therapy.
John M. Burke, MD:
Progression within the first year of initial treatment for aggressive B-cell lymphoma is associated with a poor prognosis. Second-line standard of care is typically salvage chemotherapy followed by autologous stem cell transplant. However, inadequate response to chemotherapy leaves many patients transplant-ineligible. Three randomized phase III trials presented at ASH 2021 compared CAR T‑cell therapies with the conventional second-line strategy in patients with relapsed aggressive lymphoma: BELINDA, ZUMA-7, and TRANSFORM.
First, we will discuss the BELINDA trial of tisa-cel, which enrolled 322 patients with R/R aggressive lymphoma within 12 months of their first‑line treatment.8,9 Tisa-cel is an autologous anti-CD19 CAR T-cell therapy that is approved for the treatment of patients with DLBCL after 2 or more prior lines of therapy. All patients on the BELINDA trial had to be eligible for autologous stem cell transplant. In the BELINDA study, leukapheresis was performed at screening and then patients were randomly assigned either to the control arm, which was standard-of-care platinum-based immunochemotherapy followed by autologous stem cell transplant for patients with chemosensitive disease, or to a single intravenous infusion of tisa-cel (0.6 to 6.0 x 108 CAR-positive T-cells).
The 4 platinum-based immunochemotherapy options were R plus dexamethasone/cytarabine/cisplatin (R-DHAP), R plus gemcitabine/cisplatin/dexamethasone (R-GDP), R plus ifosfamide/carboplatin/etoposide (R-ICE), or R plus gemcitabine/oxaliplatin (R-GemOx). Lymphodepletion consisted of fludarabine 25 mg/m2 plus cyclophosphamide 250 mg/m2, both given daily for 3 days. If that was contraindicated, patients could receive bendamustine 90 mg/m2/day for 2 days. Bridging chemotherapy was allowed as needed. The primary endpoint of the study was EFS.
John M. Burke, MD:
The majority of patients in each arm had DLBCL not otherwise specified. Approximately two thirds of the patients in each treatment arm had primary refractory disease, whereas the remaining patients had relapsed disease within 12 months of their initial therapy. On both arms, the median time from initial diagnosis to randomization was approximately 8 months, and the time from the most recent relapse or disease progression to randomization was just over 1 month.
John M. Burke, MD:
In the control arm, 97% of patients received 2 or more cycles of their planned platinum‑based immunochemotherapy. In the tisa-cel arm, 48% of patients received 2 or more cycles of bridging chemotherapy. The median time from leukapheresis to infusion of tisa-cel was 52 days in the overall population.
John M. Burke, MD:
There was no difference in the primary endpoint of EFS between the 2 treatment groups, with a median EFS of 3.0 months in both arms (HR: 1.07; 95% CI: 0.82-1.40; P = .69).
John M. Burke, MD:
The best overall response rate, assessed at or after Week 12, was 46.3% in the tisa-cel arm and 42.5% in the standard-of-care arm. The CR rate was similar in both arms at or after assessments at Week 12, which was 28.4% with tisa-cel and 27.5% in the standard-of-care arm.
John M. Burke, MD:
The toxicities were as expected for each group. On the tisa-cel arm, 95 patients (61.3%) experienced CRS and 16 patients (10.3%) had neurologic events. There were no new emerging concerns in the safety profiles in either of the treatment arms. Ten patients in the tisa-cel arm and 13 in the standard-of-care arm died due to adverse events.
John M. Burke, MD:
In summary, tisa-cel did not improve the primary endpoint of EFS compared with the standard approach of salvage chemotherapy followed by autologous stem cell transplant in patients with R/R aggressive lymphomas who were within 1 year of their initial therapy.
Peter Martin, MD:
The ZUMA‑7 trial is a multicenter, randomized phase III trial for patients with R/R DLBCL within 12 months of first-line immunochemotherapy. In total, 359 patients were randomized to receive axi-cel, an autologous anti-CD19 CAR T-cell therapy that is approved for patients with large B-cell lymphoma after 2 or more prior lines of therapy or standard of care, which was a platinum‑based chemoimmunotherapy regimen, selected from R-GDP, R-DHAP, R-ICE, or R plus etoposide/methylprednisolone/cytarabine/cisplatin (R-ESHAP).10,7 An important factor in this trial is that bridging therapy was limited solely to corticosteroids. No bridging chemotherapy was allowed on the ZUMA-7 trial. Perhaps a reason for this is that the manufacturing period for axi‑cel is quite short, thus reducing the need for bridging chemotherapy, but it may also have dissuaded patients in need of bridging therapy from participating in the study. The primary endpoint was EFS, and the key secondary endpoints were objective response rate (ORR) and OS.
Peter Martin, MD:
Of interest, among 179 patients who were randomized to receive standard of care, only 168 went on to receive at least 1 dose of salvage immunochemotherapy. The remaining 8 patients withdrew voluntarily from the clinical trial. Overall, 94% of the patients who were randomly assigned to the CAR T-cell arm received axi-cel, and 36% of those in the standard-of-care arm received autologous stem cell transplant.
Peter Martin, MD:
The patient characteristics were fairly balanced across both arms of the study. Overall, the majority (74%) of the patients enrolled on the ZUMA-7 trial had primary refractory disease to their first‑line immunochemotherapy. A number of patients—16%—had high‑grade B‑cell lymphoma, including double-hit and triple-hit lymphoma, and approximately one third of the patients had double‑expressor lymphoma. These characteristics are consistent with what would be expected in a high‑risk population of patients with large B-cell lymphoma.
Peter Martin, MD:
The primary endpoint of improved EFS was overwhelmingly met, with an overall EFS in the axi‑cel arm of 8.3 months (95% CI: 4.5-15.8) vs 2.0 months (95% CI: 1.6-2.8) in the standard-of-care arm (HR: 0.398; 95% CI: 0.308-0.514; P <.0001). The EFS benefit with axi-cel was seen across various subgroups regardless of age, R/R disease status, second-line age-adjusted international prognostic index category, and high-grade genomic classification of the disease. Across all these analyzed subgroups, the HR was <0.5.
Peter Martin, MD:
The ORR was also significantly higher in the axi‑cel arm (83%) compared with 50% in the standard-of-care arm, including a CR rate of 65% with axi-cel, which was approximately twice as high as that obtained with standard of care (32%).
The median OS was not reached in the axi-cel arm compared with 35.1 months in the standard-of-care arm. The HR was 0.73 (95% CI: 0.530-1.007), but currently there is no statistically significant difference in OS between arms (P = .270). However, there is a trend toward OS benefit with axi-cel over standard of care, and with longer follow-up, it will become clear if that trend is significant or not.
Peter Martin, MD:
The safety profile revealed some key differences between the 2 treatment arms. The most notable are those related to the inflammatory effects of CAR T‑cells, primarily CRS and neurologic events. The other adverse events were fairly similar between the treatment groups. There were more lymphoma-related deaths in the standard-of-care arm compared with the axi-cel arm. Overall, the treatment‑related mortality rate was similar between arms and very low in both arms. However, there was a slightly higher rate of grade 5 adverse events in the axi‑cel arm, although it is not clear if these deaths were associated with treatment.
Peter Martin, MD:
Only 6% of patients who received axi-cel experienced grade ≥3 CRS. This incidence is a bit lower than what was reported in the ZUMA‑1 trial, although it is not clear whether that is due to better prevention and management of CRS or whether it is because all patients on the ZUMA-7 trial had a slightly lower tumor burden after just 1 prior line of therapy compared with the ZUMA-1 trial where the vast majority of the patients had received at least 3 prior lines of therapy.11 On the ZUMA-7 trial, 65% of the patients received tocilizumab for CRS management, which is quite high and reflects a change in practice patterns with CAR T‑cell therapy.
Neurologic events are still a concern with CAR T-cell therapy, and on the ZUMA-7 trial, 21% of the patients in the axi‑cel arm experienced grade ≥3 neurologic events.
Peter Martin, MD:
ZUMA-7 was a positive trial. It met its primary endpoint of EFS and a key secondary endpoint of ORR. However, the interim OS analysis did not demonstrate a significant OS benefit with axi-cel. The OS endpoint may still be met with some additional follow-up. Based on the results of the ZUMA-7 trial, it appears that patients with early‑progressing large B‑cell lymphoma should be treated with axi-cel in the second‑line setting rather than standard-of-care salvage chemotherapy.
John M. Burke, MD:
I agree that this is really one of the critical results out of ASH for patients with large B‑cell lymphoma, showing that the use of axi‑cel in the second-line setting was clearly superior to conventional chemotherapy and autologous stem cell transplant. These patients experienced a rapid relapse after receiving first-line immunochemotherapy and would be predicted to have a relatively unfavorable outcome with conventional salvage therapy. The improvement in EFS with axi-cel represents a major step forward for this patient population.
Peter Martin, MD:
There are some practical questions that need to be taken into consideration, however. For example, does everyone in the United States have equal access to CAR T‑cell therapy? Can hospitals handle the inpatient volumes that might be associated with administering CAR T‑cell therapy? Can hospitals afford to give CAR T‑cells to all of these patients in the second-line setting? It is important to keep in mind that this trial included only patients with early relapsing large B-cell lymphoma, so autologous stem cell transplant may still be a treatment option, particularly in the patient population with later relapse. It is also worth noting that fewer than half of the patients (40.5%) were event free after 24 months of receiving axi-cel. We will need to wait for a longer follow-up analysis of these patients to learn how they respond to subsequent therapies, some of which target CD19, for example. As I noted earlier, there is also the question about whether the 8 patients who had been randomized to the standard-of-care arm but subsequently withdrew from the trial impacted the trial results.
After the ZUMA-7 trial results were presented in the plenary session at ASH, a question was raised about whether the study demographics, such as race and ethnicity, represent those of the population that we serve in the United States. In particular, what policies and procedures are in place to ensure that studies enroll patient cohorts that reflect the United States or global population? This is a very important question that needs to be addressed going forward in clinical trials.
John M. Burke, MD:
Similar to the BELINDA and ZUMA-7 trials, the TRANSFORM trial assessed whether CAR T‑cell therapy is more effective than salvage chemotherapy followed by autologous stem cell transplant for patients with aggressive lymphoma who experience early relapse. The TRANSFORM trial is a randomized, multicenter phase III trial that investigated lisocabtagene maraleucel (liso-cel), an autologous anti-CD19 CAR T-cell therapy that is approved for the treatment of large B-cell lymphoma after 2 or more prior lines of therapy vs standard-of-care salvage chemotherapy for patients with aggressive NHL who had experienced relapse within 12 months of their initial therapy.12
In total, 184 patients were enrolled in the TRANSFORM trial, and all patients were eligible for autologous stem cell transplant. Ninety-two patients were randomly assigned to each of the 2 arms. The standard-of-care arm consisted of 3 cycles of salvage chemotherapy followed by autologous stem cell transplant. Patients on the investigational arm of liso‑cel were given the option to receive bridging therapy. Crossover was allowed if patients on the standard-of-care arm experienced disease progression at any time or did not respond to salvage chemotherapy by 9 weeks post randomization or at the start of a new antineoplastic therapy following autologous stem cell transplant. Patients on the liso‑cel arm (n = 92) underwent lymphodepletion with fludarabine at 30 mg/m2 and cyclophosphamide at 300 mg/m2 daily for 3 days. As in the BELINDA and ZUMA-7 trials, the primary endpoint for the TRANSFORM trial was EFS as measured by an independent review committee.
John M. Burke, MD:
A little more than half of the patients on each arm had DLBCL. Almost a quarter of patients on each arm had double-hit or triple‑hit lymphoma. The majority of the patients were younger than age 65, and 73% to 74% had primary refractory disease.
John M. Burke, MD:
Out of the 92 patients who were randomly assigned to the standard-of-care arm, 1 patient withdrew consent. Therefore, 91 patients received 3 cycles of salvage chemotherapy. Only 43 patients (47%) went on to receive high-dose chemotherapy, and only 42 patients (46%) proceeded to receive autologous stem cell transplant. In the CAR T-cell arm, however, 89 out of 92 patients (97%) received liso-cel infusion.
John M. Burke, MD:
Patients who received liso‑cel achieved a median EFS of 10.1 months, whereas those who received standard-of-care chemotherapy achieved a median EFS of 2.3 months. The difference in median EFS between the 2 arms was significant (HR: 0.349; 95% CI: 0.229-0.530; P <.0001). The 12‑month EFS rates were 44.5% with liso-cel vs 23.7% with standard of care.
John M. Burke, MD:
For all analyzed patient subgroups, EFS benefit was in favor of liso-cel, regardless of the age-adjusted international prognostic index category, age group, performance status, type of NHL, DLBCL subtype, and the type of response to prior chemotherapy.
John M. Burke, MD:
Patients who received liso‑cel achieved better responses than those who received standard of care. In the liso-cel and standard-of-care arms, the CR rates were 66% and 39%, respectively, and the overall response rates were 86% and 48%, respectively.
John M. Burke, MD:
Liso‑cel prolonged PFS, with a median PFS of 14.8 months compared with 5.7 months with standard of care. The HR was 0.406 (95% CI: 0.250-0.659), which was statistically significant (P = .0001). The 12‑month PFS rates were 52.3% with liso-cel vs 33.9% with standard of care.
John M. Burke, MD:
At the time of data cutoff, the OS data were immature. The median OS was not reached in the liso‑cel group and was 16.4 months in the standard-of-care group (HR: 0.509; 95% CI: 0.258-1.004; P = .0257). There was a numerical trend in favor of liso-cel.
John M. Burke, MD:
There was 1 death in the liso‑cel group and 2 in the standard-of-care group. Overall, no new safety signals emerged in the TRANSFORM trial. The incidence of any grade neutropenia was higher with liso-cel at 82% compared with standard of care (54%), including grade ≥3 neutropenia in 80% of patients who received liso-cel vs 51% of patients who received standard of care. However, febrile neutropenia was more commonly observed with standard of care (24%) compared with liso-cel (15%).
John M. Burke, MD:
CRS (any grade) occurred in 49% of patients on the liso‑cel arm, but only 1% of these patients experienced grade ≥3 CRS. The median time to the onset of CRS was 5 days, and the median time to resolution was 4 days.
Neurologic events (any grade) were experienced by 12% of patients in the liso-cel arm, the majority of which were grade 1 or 2. The median time to the onset of neurologic events was 11 days, and the median time to resolution was 6 days. Prolonged grade ≥3 cytopenias at 35 days after liso-cel infusion occurred in 43% of patients on the CAR T-cell therapy arm, whereas only 3% of patients on the standard-of-care arm experienced prolonged cytopenias lasting for 35 days after the start of the last chemotherapy cycle. Grade ≥3 infections were slightly more frequent in the standard-of-care arm (21%) compared with the liso-cel arm (15%).
In the liso-cel arm, 26% of the patients required tocilizumab, corticosteroids, or both for the treatment of CRS and/or neurologic events. For the management of CRS alone, 23% received tocilizumab with or without steroids, whereas 8% received tocilizumab and/or steroids for the treatment of neurologic events.
Peter Martin, MD:
Three different important CAR T‑cell studies with similar patient populations but different results were presented at ASH: BELINDA, ZUMA-7, and TRANSFORM. The ZUMA‑7 and TRANSFORM trials both showed a significant improvement in EFS, whereas the BELINDA trial did not. There has been a lot of speculation regarding why different results were obtained in the different studies.
A possibility that has been proposed is there could have been a difference in the patient populations in the 3 trials, and there may have been differences in the proportion of patients on the standard-of-care arms who went on to receive autologous stem cell transplant. However, since the outcomes in the standard-of-care arms were similar, with a median EFS ranging from 2-3 months, it does not appear that the patient population in the 3 trials is the reason for the difference in the EFS outcomes with the different CAR T-cell therapies.
John M. Burke, MD:
One of the big differences between these 3 trials is the allowance of bridging chemotherapy in the TRANSFORM and the BELINDA trials, but not in the ZUMA‑7 trial. The fairly high use of bridging chemotherapy in the BELINDA trial is a major difference in the designs of the BELINDA and ZUMA‑7 trials. The prohibition of bridging chemotherapy in ZUMA‑7 may have to the selection of patients with less-bulky or advanced disease, which in turn may have allowed for the earlier delivery of CAR T-cell therapy, thereby contributing to the positive results of the trial. Of note, the 3 trials were conducted in a high‑risk group of patients, and most of the patients had primary refractory lymphoma; hence, the patients on the 3 trials do not exactly have a favorable prognosis. Of course, the TRANSFORM trial allowed bridging chemotherapy and still reported EFS benefit with CAR T‑cell therapy.
Another factor that may have contributed to the lack of EFS benefit in the BELINDA trial vs the ZUMA-7 and TRANSFORM trials is the number of salvage chemotherapy regimens allowed in the standard-of-care arms before transplant. For instance, in the BELINDA trial, patients in the standard-of-care arm with inadequate response to immunochemotherapy at 6 weeks were allowed to receive a second chemotherapy regimen to reduce tumor burden before proceeding to autologous stem cell transplant. On the ZUMA-7 trial and TRANSFORM trials, however, only 1 immunochemotherapy regimen was allowed. An additional difference in the 3 trials is the number of patients on the standard-of-care arm who proceeded to autologous stem cell transplant. After 1 regimen of salvage platinum-based immunochemotherapy, 46%, 36%, and 32.5% of patients received autologous stem cell transplant in the TRANSFORM, ZUMA-7, and BELINDA trials, respectively.
Peter Martin, MD:
It also is important to consider that there may be a difference related to the CAR T-cell product that was investigated in the different trials, perhaps due to efficacy or manufacturing time. This may have contributed to the difference in outcomes.
John M. Burke, MD:
It is not exactly clear why the results differed between these 3 studies, but the fact that we have 2 out of 3 trials demonstrating significant EFS benefit with CAR T‑cell therapy suggests that CAR T‑cell products, particularly axi‑cel and liso‑cel, have the potential to move up into the earlier treatment setting to become the standard second‑line regimen for patients with DLBCL or patients with aggressive lymphomas with early relapse or primary refractory disease.
In April 2022, axi-cel was approved for use in adult patients with large B-cell lymphoma that is refractory to first-line chemoimmunotherapy or that relapses within 12 months of first-line chemoimmunotherapy. Liso-cel has been granted priority review as second-line therapy for adults with R/R B-cell lymphoma. As these CAR T-cell products move into the second-line setting, it is necessary to understand how they will be paid for. Unfortunately, CAR T‑cell therapy is still restricted to selected centers, so not all patients who are eligible for this treatment will be able to receive it. The sooner these treatment barriers can be overcome, the better it will be for patients.
Although CAR T‑cell therapy using axi-cel and liso-cel was superior to standard of care in 2 of the 3 trials that we discussed, there is still room for improvement with these products. Even in the positive studies, a relatively high percentage of the patients experienced disease progression or died within 1 to 2 years of therapy. Overall, even though the availability of CAR T‑cell therapy is a step in the right direction, there is still a lot of work to be done to improve outcomes for more patients.
Back
