Adding Polatuzumab Vedotin to R-GemOx Extends Survival in Relapsed or Refractory Diffuse Large B-Cell Lymphoma: Results of the Phase III POLARGO Trial

Table of Contents

• Key Points
• Background: Why This Research Matters
• How the Study Was Conducted
• Who Took Part in the Study
• Key Finding: Overall Survival
• Key Finding: Progression-Free Survival and Response Rates
• Consistency Across Patient Subgroups
• Treatment Exposure and Later Therapies
• Side Effects: The Overall Picture
• Peripheral Neuropathy
• Low Blood Counts
• Infections and Deaths
• What This Means for Patients
• Limitations of the Study
• Practical Recommendations
• Frequently Asked Questions
• Source Information

Key Points

• In the phase III POLARGO trial of 255 patients with relapsed or refractory DLBCL who could not receive a stem cell transplant, Pola-R-GemOx improved median overall survival to 19.5 months versus 12.5 months with R-GemOx alone.
• The risk of death was 40% lower with Pola-R-GemOx (hazard ratio 0.60), and the risk of progression or death was 63% lower (hazard ratio 0.37) compared with R-GemOx alone.
• Complete response rates roughly doubled with Pola-R-GemOx (40.3% versus 19.0%), and overall response rates also roughly doubled (52.7% versus 24.6%).
• Peripheral neuropathy was more common with Pola-R-GemOx (57% versus 29%), though mostly mild, and fatal infections, including COVID-19, were also more frequent (11.7% versus 4.0% grade 5 events).
• The trial excluded patients with neuropathy worse than grade 1, prior polatuzumab or gemcitabine-plus-platinum, or eligibility for transplant or CAR T-cell therapy, so results may not apply to these groups.

Background: Why This Research Matters

Diffuse large B-cell lymphoma (DLBCL) is the most common aggressive lymphoma. First-line treatment works well for many patients, but outcomes drop sharply when the disease comes back (relapses) or stops responding (becomes refractory). This situation is called R/R DLBCL, short for relapsed or refractory disease.

Historically, the standard approach for R/R DLBCL was high-dose chemotherapy followed by an autologous hematopoietic stem cell transplant (HSCT) — a procedure in which a patient's own blood-forming stem cells are collected, high-dose chemotherapy is given, and the cells are then returned. Many patients, however, could not receive a transplant because of older age, poor physical fitness, other medical conditions, or disease that never responded to salvage (rescue) chemotherapy.

Treatment options have since expanded. Several CD20 × CD3 bispecific antibodies (engineered proteins that simultaneously attach to a lymphoma cell and to a patient's immune T cell, bringing them together to kill the cancer) have been approved, including glofitamab, epcoritamab, and mosunetuzumab. These agents produced high response rates and durable survival in heavily pretreated patients in early-phase pivotal studies.

Chimeric antigen receptor (CAR) T-cell therapy — a treatment in which a patient's own T cells are genetically reprogrammed to attack lymphoma — has replaced stem cell transplant in the second-line setting for patients with early relapsing or primary refractory DLBCL, as well as for some patients who cannot receive a transplant. Even so, CAR T-cell therapy has limits: access, the logistics of manufacturing and administering it, cost, and potential toxicity.

Polatuzumab vedotin is an antibody-drug conjugate. It combines an anti-CD79b monoclonal antibody (a lab-made protein that locks onto a marker called CD79b found on B cells) joined by a protease-cleavable linker — a connector that breaks apart inside the cell — to monomethyl auristatin E, a potent microtubule inhibitor (a drug that blocks the internal scaffolding cancer cells need to divide).

In the first-line setting, polatuzumab vedotin combined with rituximab, cyclophosphamide, doxorubicin, and prednisone (Pola-R-CHP) was approved after it reduced the risk of disease progression, relapse, or death by 27% compared with the older standard, R-CHOP. In the relapsed setting, polatuzumab vedotin plus bendamustine and rituximab (BR) improved the end-of-treatment complete response (CR) rate and produced an overall survival benefit.

The POLARGO trial was designed to test polatuzumab vedotin with a different chemotherapy backbone — rituximab, gemcitabine, and oxaliplatin (R-GemOx) — in patients with transplant-ineligible R/R DLBCL. The goal was to broaden treatment options and maximize clinical benefit.

How the Study Was Conducted

POLARGO (ClinicalTrials.gov identifier NCT04182204) is a multicenter, open-label, phase III trial. "Open-label" means both patients and their doctors knew which treatment was being given. The study had two parts: an initial safety run-in phase, followed by a randomized controlled trial phase.

The trial was run according to the Declaration of Helsinki, Good Clinical Practice guidelines of the International Council for Harmonisation, and applicable laws. An institutional review board or independent ethics committee at each study site approved the protocol. Every patient gave written informed consent.

Patients were recruited at 64 sites — both community and hospital centers — across 16 countries. Enrollment ran from February 7, 2020, through May 23, 2023.

Adults aged 18 years or older were eligible if they had histologically confirmed (proven by tissue examination) R/R DLBCL not otherwise specified (NOS), or a history of indolent lymphoma transforming into DLBCL. They needed to have received at least one prior line of systemic treatment.

Key exclusion criteria were:

• Active peripheral neuropathy worse than grade 1
• Prior use of polatuzumab vedotin or a gemcitabine-plus-platinum combination
• Eligibility for an autologous or allogeneic stem cell transplant or CAR T-cell therapy at the time of recruitment

After the safety run-in (15 patients treated with Pola-R-GemOx), patients were randomly assigned 1:1 to Pola-R-GemOx or R-GemOx alone. Randomization was stratified — balanced by design — according to three factors: age (70 years or younger versus older than 70), number of prior lines of systemic therapy (1 versus 2 or more), and outcome of the last systemic therapy (relapsed versus refractory).

Only the Independent Review Committee (IRC), the group that independently reviewed scans and outcomes, remained blinded to group assignments. The sponsor had no access to efficacy or safety summaries until the formal reporting of the primary analysis results.

Treatment dosing was as follows:

• Polatuzumab vedotin 1.8 mg/kg
• Rituximab 375 mg/m²
• Gemcitabine 1,000 mg/m²
• Oxaliplatin 100 mg/m²

All drugs were given intravenously every 21 days for up to eight cycles. Prophylaxis (preventive treatment) with primary granulocyte colony-stimulating factor (G-CSF, a drug that boosts white blood cell production) and anti-infective agents against Pneumocystis jirovecii pneumonia and herpes simplex virus was mandatory for all participants.

The primary endpoint was overall survival (OS) in the intention-to-treat (ITT) population — meaning every randomized patient was analyzed, regardless of whether they actually received treatment.

Key secondary endpoints included investigator-assessed progression-free survival (PFS), plus IRC-assessed complete response rate and overall response rate (ORR) at end of treatment (EOT), measured by positron emission tomography–computed tomography (PET-CT) using the Lugano criteria. Additional secondary endpoints included event-free survival (EFS), PET-CT CR and ORR at EOT, best overall response, and duration of response (DOR).

Adverse events were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0. Exploratory endpoints included biomarkers, immunogenicity, pharmacokinetics, and patient-reported outcomes captured with the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group – Neurotoxicity-12 questionnaire.

No interim analysis was planned, so the results reported here represent both the primary and the final analysis. OS and PFS were compared between groups using a two-sided log-rank test, and survival outcomes were estimated with the Kaplan-Meier method. Response rates were compared with the Cochran-Mantel-Haenszel test, with 95% confidence intervals (CIs) estimated by the Clopper-Pearson method.

Who Took Part in the Study

Of 347 patients screened, 270 were enrolled and 255 were randomly assigned. Seventy-seven patients were screen failures: 67 did not meet all eligibility criteria and 10 withdrew consent. The randomized groups were Pola-R-GemOx (129 patients), R-GemOx (126 patients), plus the 15 patients in the safety run-in.

In the ITT population, 128 patients in the Pola-R-GemOx group and 125 in the R-GemOx group received any study treatment. Baseline characteristics were well balanced between the two arms.

The median age was 65 years, with a range of 20 to 89. In the Pola-R-GemOx arm the median age was 67 years (range, 20–85); in the R-GemOx arm it was 64 years (range, 24–89). Patients aged 70 or younger made up 65.1% of each arm (84 and 82 patients; 166 of 255 overall), while those older than 70 made up 34.9% of each arm (45 and 44 patients; 89 overall).

Men numbered 73 (56.6%) in the Pola-R-GemOx arm and 61 (48.4%) in the R-GemOx arm — 134 patients, or 52.5%, overall. Women numbered 56 (43.4%) and 65 (51.6%) — 121 patients, or 47.5%, overall.

Race distribution across the two arms was:

• American Indian or Alaska Native: 7 (5.4%) and 9 (7.1%) — 16 patients, 6.3% overall
• Asian: 49 (38.0%) and 41 (32.5%) — 90 patients, 35.3% overall
• Black or African American: 3 (2.3%) and 6 (4.8%) — 9 patients, 3.5% overall
• White: 60 (46.5%) and 56 (44.4%) — 116 patients, 45.5% overall
• Other: 8 (6.2%) and 6 (4.8%) — 14 patients, 5.5% overall
• Unknown: 2 (1.6%) and 8 (6.3%) — 10 patients, 3.9% overall

Hispanic or Latino ethnicity was reported by 27 patients (20.9%) in the Pola-R-GemOx arm and 26 (20.6%) in the R-GemOx arm — 53 patients, 20.8% overall. Non-Hispanic or Latino ethnicity applied to 97 (75.2%) and 90 (71.4%) patients — 187, or 73.3% overall. Ethnicity was not reported for 5 (3.9%) and 10 (7.9%) patients.

Most patients were relatively fit: 225 of 255 (88.2%) had an Eastern Cooperative Oncology Group performance status (ECOG PS, a 0–5 scale of daily-functioning ability) of 0 or 1, while 30 (11.8%) scored 2. Advanced disease was common — 195 patients (76.5%) had Ann Arbor stage III or IV disease, versus 60 (23.5%) with stage I or II.

The International Prognostic Index (IPI, a risk score for lymphoma) was 2 or lower in 129 patients (50.6%) and 3 to 5 in 126 patients (49.4%).

Nearly all patients — 232 (91.0%) — had DLBCL not otherwise specified, while 23 (9.0%) had a history of transformation from indolent disease. Importantly, all patients initially had a DLBCL diagnosis confirmed by the investigator and site pathologist, but after enrollment, central pathology review confirmed that 10 patients actually had double-hit lymphoma (a high-risk subtype with two genetic rearrangements).

Most patients had received only one prior line of therapy: 165 of 255 (64.7%). The remaining 90 (35.3%) had received two or more. The median number of prior lines was 1, with a range of 1 to 6 in the Pola-R-GemOx arm and 1 to 7 in the R-GemOx arm.

Disease was refractory to the last prior treatment in 168 patients (65.9%), and relapsed in 87 (34.1%). Among the 232 patients with DLBCL NOS, 136 (58.6%) had been refractory to their first treatment and were therefore classified as primary refractory; 96 (41.4%) were not.

Prior treatments included stem cell transplant in 13 patients overall (5 in the Pola-R-GemOx arm [3.9%] and 8 in the R-GemOx arm [6.3%]), and CAR T-cell therapy in 1 patient (0.4%, in the R-GemOx arm). No patients had received prior bispecific antibodies.

Bulky disease (a mass of at least 7.5 cm) was present in 48 patients overall (18.8%) — 23 (17.8%) and 25 (19.8%) in the respective arms. Cell of origin, determined centrally, was activated B-cell–like (ABC) in 86 patients (33.7%), germinal center B-cell–like (GCB) in 98 (38.4%), unclassified in 20 (7.8%), and unknown in 51 (20.0%).

The most common reason patients could not receive a stem cell transplant was age, cited for 104 patients (40.8%).

Key Finding: Overall Survival

The study met its primary endpoint. After a median follow-up of 24.6 months (range, 0–34 months), patients receiving Pola-R-GemOx had a statistically significant 40% lower risk of death than those receiving R-GemOx.

• Hazard ratio (HR): 0.60 (95% CI, 0.43 to 0.83); p=.0017. A hazard ratio below 1.0 means the treatment group had a lower risk of the event; 0.60 means a 40% relative reduction in the risk of death.
• Median overall survival: 19.5 months with Pola-R-GemOx (95% CI, 13.3 months to not estimable) versus 12.5 months with R-GemOx (95% CI, 8.9 to 15.8 months). "Not estimable" means the upper boundary of the range had not yet been reached when the data were analyzed.
• 12-month survival rate: 64.5% (95% CI, 56.1% to 72.8%) versus 50.2% (95% CI, 41.2% to 59.1%). In plain terms, about 6 in 10 patients were alive at one year on the polatuzumab regimen, compared with about 5 in 10 on R-GemOx alone.
• 24-month survival rate: 44.0% (95% CI, 34.9% to 53.2%) versus 33.2% (95% CI, 24.5% to 41.8%) — roughly 4 in 10 versus 3 in 10.

The p-value of .0017 means there is about a 0.17% probability that a difference this large arose by chance alone. Fewer deaths were attributed to progressive disease (PD, meaning the lymphoma growing or spreading) in the Pola-R-GemOx arm: 38 patients (29.7%) versus 70 patients (56.0%) in the R-GemOx arm.

Key Finding: Progression-Free Survival and Response Rates

Every key secondary endpoint favored Pola-R-GemOx, and all reached statistical significance.

Progression-free survival (investigator-assessed): the risk of disease progression or death fell by 63% with Pola-R-GemOx (HR, 0.37; 95% CI, 0.27 to 0.51; p<.0001).

• Median PFS: 7.4 months (95% CI, 6.2 to 11.7) versus 2.7 months (95% CI, 2.4 to 3.3)
• 12-month PFS rate: 36.6% (95% CI, 27.6% to 45.5%) versus 17.9% (95% CI, 10.3% to 25.5%)
• 24-month PFS rate: 25.3% (95% CI, 16.7% to 33.9%) versus 11.5% (95% CI, 4.2% to 18.7%)

Complete response (CR) at end of treatment, assessed by the independent review committee using PET-CT: 40.3% with Pola-R-GemOx (52 of 129 patients; 95% CI, 31.8% to 49.3%) versus 19.0% with R-GemOx (24 of 126 patients; 95% CI, 12.6% to 27.0%). This difference was statistically significant (p<.0001). A complete response means no detectable evidence of lymphoma on imaging.

Overall response rate at end of treatment: 52.7% with Pola-R-GemOx (68 responders; 95% CI, 43.7% to 61.6%) versus 24.6% with R-GemOx (31 responders; 95% CI, 17.4% to 33.1%), also highly significant (p<.0001). The overall response rate includes both complete and partial responses.

Reassuringly, investigator and independent review committee assessments agreed closely with one another: 94% concordance for complete response and 88% concordance for overall response. Additional secondary endpoints are reported in the article's data supplement.

Consistency Across Patient Subgroups

The overall survival benefit of Pola-R-GemOx held up across most prespecified subgroups. This is important because it suggests the benefit is not limited to one narrow group of patients. The advantage was seen in:

• Patients with primary refractory disease
• Patients receiving the regimen as second-line treatment for DLBCL
• Patients with both activated B-cell–like (ABC) and germinal center B-cell–like (GCB) cell-of-origin subtypes
• Patients with double-expressor DLBCL (a subtype in which two specific proteins are overproduced, determined by central immunohistochemistry testing)

Treatment Exposure and Later Therapies

Patients on Pola-R-GemOx stayed on treatment longer. They received a median of 7.5 cycles, compared with 4.0 cycles in the R-GemOx arm.

Half of the Pola-R-GemOx group — 64 patients (50.0%) — completed all eight planned cycles, versus 29 patients (23.4%) in the R-GemOx arm. Importantly, adding polatuzumab vedotin did not compromise the delivery of the chemotherapy backbone: the median dose intensity was 96.2% with Pola-R-GemOx and 98.6% with R-GemOx.

Fewer patients on Pola-R-GemOx went on to need another therapy afterwards: 59 patients (45.7%) versus 82 (65.1%) in the R-GemOx arm. Use of subsequent T-cell–engaging therapies (CAR T-cell therapy, bispecific regimens, or stem cell transplant) was balanced between arms and did not substantially affect the efficacy results. Most patients received subsequent therapy after their disease progressed.

Side Effects: The Overall Picture

The safety analysis included 128 patients in the Pola-R-GemOx arm and 125 in the R-GemOx arm. Overall rates of side effects were broadly similar between groups, but a few categories differed.

Treatment-emergent adverse events (AEs, meaning any unwanted medical occurrence during treatment) of any grade occurred in 125 patients (97.7%) on Pola-R-GemOx and 117 patients (93.6%) on R-GemOx. Grade 3 or 4 AEs (moderate-to-severe events requiring medical intervention) occurred in 73 patients (57.0%) and 73 patients (58.4%), respectively.

Serious adverse events occurred in 49 patients (38.3%) on Pola-R-GemOx and 39 patients (31.2%) on R-GemOx.

Grade 5 adverse events — events that resulted in death — were more frequent with Pola-R-GemOx: 15 patients (11.7%) versus 5 patients (4.0%). This difference was largely driven by infections, including COVID-19. Fatal COVID-19 events occurred in seven patients (5.5%) receiving Pola-R-GemOx versus two patients (1.6%) receiving R-GemOx.

Importantly, when investigators judged whether a fatal event was caused by treatment, the rates were nearly the same: four patients (3.1%) on Pola-R-GemOx and three patients (2.4%) on R-GemOx.

Deaths occurred in 68 patients (53.1%) in the Pola-R-GemOx arm and 83 patients (66.4%) in the R-GemOx arm. Causes broke down as follows:

• Progressive disease: 38 patients (29.7%) versus 70 patients (56.0%)
• Grade 5 adverse events: 15 patients (11.7%) versus 5 patients (4.0%)
• Deaths not due to progressive disease occurring outside the adverse event reporting period: 15 patients (11.7%) versus 8 patients (6.4%)

Treatment discontinuation differed between arms. In the R-GemOx arm, adverse events led to stopping at least one study drug in 10 patients (8.0%). In the Pola-R-GemOx arm, 30 patients (23.4%) discontinued a study drug because of toxicity. The most common reasons for discontinuation were infections (9 patients [7.0%] versus 2 patients [1.6%]) and low platelet counts (7 patients [5.5%] versus 4 patients [3.2%]).

In the article's commentary, JCO Editor-in-Chief Jonathan W. Friedberg, MD, noted that chemotherapy with rituximab alone is now outdated for relapsed DLBCL, because combinations including polatuzumab (such as this trial's regimen) or bispecific antibodies demonstrate improved survival. He suggested this combination may be an appropriate salvage option before definitive CAR T-cell therapy, and noted that ongoing studies are evaluating polatuzumab plus bispecific antibody combinations, with or without chemotherapy, in both frontline and relapsed settings.

Peripheral Neuropathy

Peripheral neuropathy (nerve damage that can cause numbness, tingling, burning, or weakness, usually starting in the hands or feet) was the most notable side effect difference between the arms. It occurred in 73 patients (57%) on Pola-R-GemOx versus 36 patients (29%) on R-GemOx.

The reassuring news is that most of it was mild:

• Grade 1 neuropathy (mild symptoms): 48 patients (37.5%) versus 29 patients (23.2%)
• Grade 2 neuropathy (moderate symptoms): 20 patients (15.6%) versus 7 patients (5.6%)

In other words, the excess neuropathy with Pola-R-GemOx was primarily grade 1 — often described as numbness or tingling that does not interfere with daily activities. Patients with pre-existing peripheral neuropathy worse than grade 1 were excluded from the trial, so this regimen's neuropathy risk in that population is unknown.

Low Blood Counts

Blood count problems were the most common serious laboratory side effects in both arms, though the patterns differed slightly.

Low platelets (thrombocytopenia): any grade in 68 patients (53.1%) on Pola-R-GemOx versus 51 patients (40.8%) on R-GemOx. Grade 3 or 4 (severe) in 44 patients (34.4%) versus 33 patients (26.4%). Platelet transfusions were needed in 12 patients (9.4%) versus 16 patients (12.8%).

Low red blood cells (anemia): any grade in 48 patients (37.5%) versus 35 patients (28.0%). Grade 3 or 4 in 17 patients (13.3%) versus 19 patients (15.2%). Red blood cell transfusions were given to 18 patients (14.1%) versus 22 patients (17.6%).

Low neutrophils (neutropenia), a type of white blood cell that fights infection: rates were similar in both arms — any grade in 53 patients (41.4%) versus 52 patients (41.6%), and grade 3 or 4 in 43 patients (33.6%) versus 38 patients (30.4%).

Febrile neutropenia (a fever with low neutrophil counts, which can signal a dangerous infection) was uncommon, occurring in just three patients in each arm.

Nearly all patients received preventive G-CSF: 116 patients (90.6%) in the Pola-R-GemOx arm and 114 patients (91.2%) in the R-GemOx arm.

Infections and Deaths

Infections and infestations increased with the addition of polatuzumab vedotin. This is the single most important safety signal from the trial. Fatal infections drove most of the difference in deaths from adverse events, and COVID-19 was a prominent contributor.

This finding deserves context. The trial enrolled patients between February 2020 and May 2023 — spanning the height of the COVID-19 pandemic. Patients with lymphoma who receive B-cell–directed therapies such as rituximab and polatuzumab vedotin are known to mount weaker responses to infection and to vaccination.

Because of this, the trial required mandatory prophylaxis against Pneumocystis jirovecii pneumonia and herpes simplex virus, along with routine G-CSF support. Even so, the infection risk remained higher on the polatuzumab-containing arm, and patients considering this regimen today should discuss vaccination status, infection prevention, and monitoring with their care team.

What This Means for Patients

For patients with relapsed or refractory DLBCL who cannot receive a stem cell transplant, Pola-R-GemOx offers a meaningful survival advantage over R-GemOx alone. The numbers tell a clear story:

• Median overall survival improved by 7 months (19.5 versus 12.5 months)
• The relative risk of death dropped by 40% (HR 0.60)
• The relative risk of progression or death dropped by 63% (HR 0.37)
• Complete response rates roughly doubled (40.3% versus 19.0%)
• Overall response rates roughly doubled (52.7% versus 24.6%)

Just as important, the benefit appeared consistent across several patient types — including those with primary refractory disease, those receiving only their second line of therapy, and those with either ABC or GCB cell-of-origin subtypes. That breadth matters, because DLBCL is not one uniform disease.

The trade-off is tolerability. Peripheral neuropathy is considerably more common with the polatuzumab-containing regimen, although usually mild. Fatal infections — particularly COVID-19 — were also more frequent. Patients and clinicians must weigh the survival gain against these risks together.

It is also worth noting the wider treatment landscape. As the journal commentary emphasized, options for relapsed DLBCL have broadened considerably, including bispecific antibodies such as glofitamab, epcoritamab, and mosunetuzumab, and CAR T-cell therapy. Pola-R-GemOx may serve as an effective salvage option, potentially used before definitive CAR T-cell therapy.

Limitations of the Study

Several factors limit what this trial can tell us.

1. The study was open-label. Patients and investigators knew which treatment was being given. This can influence how side effects are reported and how investigators assess response. The trial addressed part of this by keeping the Independent Review Committee blinded to assignments.
2. Enrollment spanned the COVID-19 pandemic. Fatal infections, especially COVID-19, were a major driver of the difference in deaths from adverse events. The infection risk profile may look different now with widespread vaccination and available antiviral treatments.
3. Certain patients were excluded. Those with peripheral neuropathy worse than grade 1, prior polatuzumab vedotin exposure, or prior gemcitabine-plus-platinum treatment could not enroll. Patients eligible for transplant or CAR T-cell therapy were also excluded. Results may not apply to these groups.
4. A small number of participants were misclassified at enrollment. Ten patients were later found by central pathology review to have double-hit lymphoma, a higher-risk subtype, meaning the trial population was slightly more heterogeneous than planned.
5. No interim analysis was planned, and these results represent both the primary and final analysis. The median follow-up of 24.6 months (range, 0–34 months) means longer-term outcomes remain to be seen.
6. The regimen was tested against R-GemOx, not against newer agents such as bispecific antibodies or CAR T-cell therapy. So this trial cannot say whether Pola-R-GemOx is better or worse than those options. The commentary noted that ongoing studies are evaluating polatuzumab plus bispecific antibody combinations in both frontline and relapsed settings.

Practical Recommendations

If you or a loved one has relapsed or refractory DLBCL and cannot receive a stem cell transplant, here are concrete steps to consider.

1. Ask whether Pola-R-GemOx is an option for you. Discuss your transplant eligibility, prior treatments (especially any prior polatuzumab, gemcitabine, or platinum-based therapy), and any existing nerve symptoms.
2. Report nerve symptoms early. Because peripheral neuropathy is more common with this regimen, tell your care team about any new numbness, tingling, burning, or weakness in your hands or feet right away. Dose adjustments are often possible, and early reporting helps prevent it from worsening.
3. Make sure infection prevention is in place. Confirm that you will receive the required prophylaxis against Pneumocystis jirovecii pneumonia and herpes simplex virus, as well as G-CSF support.
4. Review your vaccination status — including COVID-19 — with your care team before and during treatment, since B-cell–directed therapies like rituximab and polatuzumab vedotin weaken immune responses to infection and vaccination.
5. Expect close blood count monitoring. Low platelets, red blood cells, and neutrophils were common. Your team will track these with regular blood tests, and transfusions are available if needed.
6. Plan for up to eight cycles every 21 days, since half of patients in the trial completed all eight cycles. Ask what the plan is if you develop side effects, and how your team will decide whether to continue.
7. Discuss the whole treatment sequence. Ask how Pola-R-GemOx fits with other options such as bispecific antibodies and CAR T-cell therapy, and whether it might serve as a bridge to a later, definitive therapy.

Frequently Asked Questions

What is Pola-R-GemOx and who was it tested in?

Pola-R-GemOx combines polatuzumab vedotin with rituximab, gemcitabine, and oxaliplatin. In the phase III POLARGO trial, it was tested in 255 adults with relapsed or refractory diffuse large B-cell lymphoma who could not receive a stem cell transplant. Patients had received at least one prior treatment, and most had only one prior line of therapy.

How much did Pola-R-GemOx extend survival compared with R-GemOx alone?

In the POLARGO trial of 255 patients, median overall survival was 19.5 months with Pola-R-GemOx versus 12.5 months with R-GemOx alone. That is a 40% lower risk of death (hazard ratio 0.60). About 6 in 10 patients were alive at one year on the polatuzumab regimen, compared with about 5 in 10 on R-GemOx alone.

What does a hazard ratio of 0.60 mean for my treatment?

A hazard ratio below 1.0 means the treatment group had a lower risk of the event. In the POLARGO trial, a hazard ratio of 0.60 for overall survival means patients receiving Pola-R-GemOx had a 40% lower risk of death than those receiving R-GemOx alone. This result was statistically significant, with a p-value of .0017.

What side effects were more common with Pola-R-GemOx?

In the POLARGO trial, peripheral neuropathy (nerve damage causing numbness or tingling) occurred in 57% of patients on Pola-R-GemOx versus 29% on R-GemOx, but most cases were mild (grade 1). Fatal infections, including COVID-19, were also more frequent with Pola-R-GemOx (11.7% versus 4.0% grade 5 events). Low blood counts occurred in both groups.

How long is Pola-R-GemOx given, and how many cycles did patients complete?

In the POLARGO trial, all drugs were given intravenously every 21 days for up to eight cycles. Patients on Pola-R-GemOx received a median of 7.5 cycles, and half (50.0%) completed all eight planned cycles, compared with 23.4% on R-GemOx alone. Adding polatuzumab did not compromise delivery of the chemotherapy backbone.

Who was not eligible for the POLARGO trial?

The POLARGO trial excluded patients with active peripheral neuropathy worse than grade 1, prior use of polatuzumab vedotin or a gemcitabine-plus-platinum combination, and those eligible for a stem cell transplant or CAR T-cell therapy at the time of recruitment. Results may not apply to these groups. Always discuss your own eligibility with your care team.

What should I ask my doctor about infection prevention during Pola-R-GemOx?

In the POLARGO trial, mandatory preventive treatment included G-CSF to boost white blood cells and anti-infective agents against Pneumocystis jirovecii pneumonia and herpes simplex virus. Even so, infections were more common with Pola-R-GemOx. Ask your care team about your vaccination status, including COVID-19, and monitoring, because B-cell–directed therapies can weaken immune responses.

I have relapsed DLBCL and can't have a stem cell transplant — should I get a second opinion before starting Pola-R-GemOx?

A second opinion can be valuable here because the decision involves weighing a real survival gain against added toxicity. Pola-R-GemOx extended median overall survival to 19.5 months versus 12.5 months with R-GemOx, and roughly doubled complete response rates, but peripheral neuropathy was more common and fatal infections, including COVID-19, were higher. An independent review can confirm your transplant ineligibility, prior treatments, and nerve symptoms, and help place Pola-R-GemOx alongside bispecific antibodies and CAR T-cell therapy in your sequence. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Polatuzumab Vedotin Plus Rituximab, Gemcitabine, and Oxaliplatin in Relapsed or Refractory Diffuse Large B-Cell Lymphoma: Results From the Phase III, Randomized POLARGO Trial.

Authors: Matasar M, Li Z, Vassilakopoulos TP, Sancho JM, Viardot A, McMillan A, Dal MS, Pereira J, Kim JS, Qiu L, Batlevi CL, Ibrahim R, Hernandez J, McCall B, Jiang Y, Yan M, Harris W, Musick L, Haioun C.

Publication details: Journal of Clinical Oncology (J Clin Oncol), volume 44, issue 25, pages 2435–2446. Accepted April 30, 2026; published July 6, 2026. © 2026 by American Society of Clinical Oncology. DOI: https://doi.org/10.1200/JCO-25-02849

Trial registration: POLARGO, ClinicalTrials.gov identifier NCT04182204

Note: This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and does not replace discussion with your own oncology care team. Treatment decisions should always be individualized with your physicians and other health care providers.