Polatuzumab vedotin–based therapy as third line or beyond treatment for patients with DLBCL

Polatuzumab vedotin—a CD79b-targeted antibody-drug conjugate—has recently shown promising activity when combined with bendamustine-rituximab in patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL). However, the research describing prognostic factors for patients receiving polatuzumab-based chemotherapy is limited. Additionally, a majority of the available studies enrolled patients who were ineligible for hematopoietic stem cell transplantation (HSCT) and therefore it is unknown if polatuzumab has any benefit in patients who are eligible for HSCT. In a retrospective study, Yu‑Wen Wang, et al.¹ evaluated the efficacy and safety of polatuzumab-based chemotherapy in the third-line and beyond as treatment for patients with R/R DLBCL and aimed to identify prognostic factors in this population. This study (NCT05006534) also explored the role of HSCT in a high-risk population following treatment with polatuzumab-based chemotherapy.

Study design

Baseline characteristics

In total, 40 patients with R/R DLBCL (92.5% with de novo and 7.5% with transformed lymphoma) who underwent salvage therapy with a polatuzumab-containing regimen for at least one cycle between November 2018 and April 2021 were enrolled in this retrospective observational study. Patient characteristics are shown in Table 1.

Table 1. Baseline characteristics*

ALC, absolute lymphocyte count; AMC, absolute monocyte count; ANC, absolute neutrophil count; GCB, germinal center B-cell; HBV, hepatitis B virus; HSCT, hematopoietic stem cell transplantation; PS, prognostic score; R-IPI, revised International Prognostic Index. *Adapted from Wang, et al.1
Characteristics	N = 40
Median age (range), years	59 (20–82)
Female, %	55
Cell of origin, %
GCB	40
Non-GCB	52.5
Undetermined	7.5
Prior treatment lines, %
2	20
3	7.5
4	37.5
5–11	35
Ann Arbor stages III–IV, %	77.5
R-IPI risk stratification at diagnosis, %
Low	5
Intermediate	17.5
High	70
Unknown	7.5
R-IPI risk stratification before polatuzumab vedotin, %
Low	7.5
Intermediate	22.5
High	70
ANC (106/µL) (range), median	3,580 (499–12,310)
ALC (106/µL) (range), median	681 (0–3,008)
AMC (106/µL) (range), median	446 (53–2,156)
ANC/ALC	4.53 (0.37‒>100)
ALC/AMC	1.40 (<0.01‒14.22)
ALC/AMC PS before polatuzumab vedotin, %
Low	512.5
Intermediate	65
High	27.5
LDH (IU/mL) (range), median	308 (108–2,789)
Prior autologous HSCT, %	22.5
Polatuzumab vedotin monotherapy, %	2.5
Polatuzumab vedotin combination regimen, %	97.5
Rituximab	7.5
Rituximab + bendamustine	52.5
Rituximab + gemcitabine-based	15
Rituximab + carmustine-based	10
Others, %	12.5
Cycles of polatuzumab vedotin treatment (range), median	3 (1–6)

The 1.8 mg/kg polatuzumab vedotin dose was administered based on actual body weight, and all patients received tumor lysis syndrome (TLS) prophylaxis (hydration and febuxostat). Granulocyte colony-stimulating factor was given prophylactically for neutropenia at the attending physician’s discretion.

Treatment response was determined using positron emission tomography/computed tomography according to the Lugano criteria and defined as follows:

• Overall survival (OS): time from receipt of polatuzumab vedotin-based treatment to death regardless of cause.
• Progression free survival: time from receipt of polatuzumab vedotin-based treatment to relapse (for patients achieving complete response [CR]) or progression (for patients achieving partial remission [PR]).

The Common Terminology Criteria for Adverse Events version 5.0 was used to grade treatment-related adverse events.

Results

The median follow-up was 18.8 months, and the median OS of the total cohort was 8.5 months, with an estimated one-year OS of 35.9%. The median OS was significantly improved in patients who achieved CR or PR compared to those who did not (24.0 months vs 4.2 months, p < 0.001), and the median progression free survival of patients who achieved a CR or PR was 30.2 months. Patients who received a subsequent HSCT had a median OS of 24 months.

 

The overall response rate (ORR) based on the Lugano criteria was 52.5%, including CR and PR rates of 25.0% and 27.5%, respectively. Disease progression was observed in 45.0% of patients, and one death due to infection was recorded before disease evaluation. Patients with non-germinal center B-cell DLBCL had a higher ORR than those with germinal center B-cell DLBCL (65.0% vs 31.3%; p = 0.044), though there was no survival benefit between these groups (9.8 months vs 5.6 months, respectively; p = 0.267).

The revised International Prognostic Index (R-IPI) risk stratification was sufficient to predict OS at diagnosis and at the time of polatuzumab treatment:

• Low/intermediate R-IPI risk score at diagnosis and prior to polatuzumab treatment predicted better OS: patients who had R-IPI scores ≥3 before polatuzumab-based salvage treatment had a lower ORR (40.7% vs 83.3%; p = 0.014).

• Patients with low/intermediate risk at diagnosis had a median OS of not reached, while patients with high risk at diagnosis had a median OS of 6.2 months (p = 0.008)
• Patients with low/intermediate risk at time of polatuzumab treatment had a median OS of 24 months, while patients with high risk at time of treatment had a median OS of 5.6 months (p = 0.005)

Regarding clinical characteristics as predictors of response:

• Patients with lactate dehydrogenase (LDH) within the reference range prior to polatuzumab treatment had a higher ORR (68.4% vs 40.0%; p = 0.075).
• Elevated LDH before polatuzumab administration was a poor prognostic factor for OS (median, 5.4 months vs 24.0 months; p = 0.009).
• The absolute lymphocyte count (ALC)/absolute monocyte count (AMC) prognostic score was able to effectively predict the ORR in different risk groups (p = 0.027):
  • High risk, 22.2%
  • Intermediate risk, 57.8%
  • Low risk, 100.0%
• Patients with an ALC/AMC ratio >1.5 had significantly greater OS compared to those with a ratio ≤1.5 (median, 24 months vs 5.6 months; p = 0.014).
• Disease type, ALC, absolute neutrophil count (ANC)/ALC ratio, and previous HSCT were not statistically significant prognostic factors.
• There was no significant survival difference in patients in the low-, intermediate-, and high-risk groups according to the ALC/AMC prognostic score (9.8 months vs 6.5 months vs 5.4 months; p = 0.203).

A total of 32.5% of patients bridged to autologous or allogeneic HSCT:

• OS was improved significantly in patients who bridged to HSCT (24.0 months vs. 4.4 months; p = 0.001).
• Patients who reached CR or PR after polatuzumab-based treatment and received subsequent HSCT had significantly better OS (24.0 months vs 5.4 months; p = 0.026).
• Polatuzumab-based treatment yielded a poor response in one patient; therefore, they received salvage treatment before HSCT.

There was no difference in OS between patients who had >4 prior lines of therapy and patients who were not heavily pre-treated, or between those who had de novo DLBCL vs transformed lymphoma.

Three of the five patients with secondary central nervous system lymphoma prior to receiving polatuzumab-based treatment achieved PR.

Safety

The most common adverse event of any grade was cytopenia. The most common hematologic adverse events were neutropenia (87.5%), which led to granulocyte colony-stimulating factor support in all patients who were receiving polatuzumab-based therapy, thrombocytopenia (95.0%),

which lead to a dose reduction of polatuzumab vedotin in one patient (who experienced Grade 4 thrombocytopenia), and anemia (90.0%).

Peripheral neuropathy associated with polatuzumab vedotin was observed in 25% of patients, and four patients experienced Grade 3 peripheral neuropathy. Infections of any grade were reported in 50% of patients. Infections (shown in Table 2) resulted in three deaths in patients achieving an overall response.

Table 2. Safety profile of polatuzumab vedotin-based therapy*

CMV, cytomegalovirus; GPC, gram-positive cocci; GNB, gram-negative bacilli; HBV, hepatitis B virus. *Adapted from Wang, et al.1 †Two patients had vancomycin-resistant Enterococcus faecium bacteremia, one patient had Salmonella spp. bacteremia, and one patient had concurrent Escherichia coli and Klebsiella pneumoniae bacteremia. ‡One patient had Acinetobacter baumannii and Pseudomonas aeruginosa coinfection, one patient had Acinetobacter ursingii pneumonia, and one patient had vancomycin-resistant Enterococcus faecium and Candida parapsilosis coinfection.
Adverse events, n (%)	All grades	Grade 3–4
Tumor lysis syndrome, clinical	2 (5)	2 (5)
Anemia	36 (90)	22 (55)
Neutropenia	35 (87.5)	26 (65)
Received at least 1               dose of growth factor	40 (100)	—
Febrile neutropenia	17 (42.5)	—
Thrombocytopenia	38 (95)	28 (70)
Documented infection during treatment	12 (30)	—
Bacteremia/GPC/GNB†	4 (10)	—
Bacterial pneumonia‡	3 (7.5)	—
Candida tropicalis               fungemia	1 (2.5)	—
Invasive fungal               infection	5 (12.5)	—
Viral reactivation during follow-up		—
CMV	1 (2.5)	—
HBV	3 (7.5)	—
Peripheral neuropathy	10 (25)	4 (10)
30-day early mortality	2 (5)	—

Conclusion

In conclusion, this study shows that polatuzumab vedotin combined with either bendamustine-rituximab or other intensive chemotherapy is a well-tolerated and promising therapy option for patients with R/R DLBCL. Treatment benefits were reported in patients with LDH levels within the normal range, low- or intermediate-risk R-IPI scores, and/or ALC/AMC ratios >1.5. However, there were safety concerns regarding the high frequency of reported cytopenia and infections; with the three reported deaths, the cumulative 30-day mortality rate was 5%. Further studies of this monoclonal antibody–drug conjugate are expected in future clinical trials for DLBCL.