Efficacy and safety of Pola-R-mini-CHP therapy in patients aged ≥80 years with CD5-positive diffuse large B-cell lymphoma

SHORT COMMUNICATION

CD5-positive diffuse large B-cell lymphoma (DLBCL) accounts for 5%–22% of all cases of DLBCL, not otherwise specified.^1–3 Approximately 80% of these cases are of the activated B-cell subtype.¹

Patients with CD5-positive DLBCL are often older women and frequently present with B symptoms, advanced disease stage at diagnosis, elevated serum lactate dehydrogenase levels, extranodal involvement, an Eastern Cooperative Oncology Group performance status score of >1, and an International Prognostic Index (IPI) score of >2. Additionally, these patients commonly exhibit bone marrow and central nervous system (CNS) infiltration.^1–3 Double-expressor (DE) status and tumor cell P53 expression are also frequently observed.^2,4 Patient prognosis and response to R-CHOP therapy (rituximab + cyclophosphamide + doxorubicin + vincristine + prednisolone) are generally poor in this population.^5–8

A previous study found that dose-adjusted EPOCH-R (DA-EPOCH-R; etoposide + prednisolone + vincristine + cyclophosphamide + doxorubicin + rituximab) therapy improved progression-free survival (PFS); however, overall survival (OS) did not differ significantly from that achieved with R-CHOP therapy.⁹ Moreover, the reported median OS following DA-EPOCH-R therapy remains poor (28.13 months),¹⁰ and its potential to improve prognosis of patients with CD5-positive DLBCL remains unclear.²

Two phase II clinical trials evaluating DA-EPOCH-R plus high-dose methotrexate (HD-MTX) have been reported. After a median follow-up of 3.1 years, favorable outcomes were observed, including a CNS recurrence rate of 9%, 2-year PFS of 79%, and 2-year OS of 89%.¹¹ Subsequently, a multicenter retrospective observational study reported a 2-year PFS of 76%, 2-year OS of 87%, and 2-year cumulative CNS recurrence rate of 7.3% after a median follow-up of 43 months. However, the cumulative CNS recurrence rate in patients receiving prophylactic HD-MTX did not differ significantly from that in patients who did not receive it.¹ Thus, the optimal treatment strategy for CD5-positive DLBCL remains unknown.

Furthermore, CD5 positivity has been identified as an independent adverse prognostic factor in older patients with DLBCL.⁵

We previously reported the efficacy and safety of Pola-R-mini-CHP therapy (rituximab 375 mg/m², cyclophosphamide 400 mg/m², doxorubicin 25 mg/m², polatuzumab vedotin 1.8 mg/kg on day 1 of each cycle, and prednisolone 40 mg/m² on days 1–5 of each cycle, administered every 3 weeks) in patients with DLBCL aged ≥80 years.^12,13 The present study aimed to further evaluate the efficacy and safety of this regimen, focusing on patients aged ≥80 years with CD5-positive DLBCL.

PATIENTS AND METHODS

We conducted a retrospective analysis of six patients aged ≥80 years with CD5-positive DLBCL who received Pola-R-mini-CHP therapy at Saitama Cancer Center between March 2023 and February 2025. Treatment response was evaluated by reassessing findings from computed tomography and/or positron emission tomography–computed tomography according to the Deauville criteria of the Lugano classification.¹⁴ Hematological and non-hematological toxicities observed during treatment were graded according to the National Cancer Institute Common Toxicity Criteria, version 5.0. This study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol and related documents were approved by the Ethics Committee of Saitama Cancer Center on September 4, 2023, and continuation of the study was approved by the same committee on October 31, 2025. Informed consent was obtained using an opt-out method via the institutional website.

RESULTS

Patient characteristics are summarized in Table 1. The median patient age was 83.5 (81–84) years, and all patients had disease stage ≥III. The IPI score was ≥ 3 in three patients, and the CNS-IPI score was ≥ 4 in three patients. The germinal center B-cell (GCB) subtype was observed in four patients. No cases of double-hit lymphoma were identified, whereas three patients had DE lymphoma. All patients received sulfamethoxazole/trimethoprim, tixagevimab–cilgavimab, and pegfilgrastim as prophylactic therapy. Acyclovir was administered at the discretion of the treating physician. Although the GCB subtype was the most common (n = 4), the reason for this predominance remains unclear. No differences in clinical characteristics were observed between patients with GCB and non-GCB DLBCL in this cohort.

Table 1. Patient characteristics.

Case	Age/Sex	PS	Stage	LDH	EN	IPI	CNS IPI	Status	IHC COO	CD5 Expression (IHC)	BCL2/MYC Rearrangement (FISH)	BCL2/MYC Expresser (IHC)	CCI	Treatment Progress	Reason for discontinuation	Response	Outcome
1	82/F	0	III	179	0	2	2	De novo	CD10(-) bcl-6(+) MUM-1(+) non-GCB	> 95%	-	-	4	8C end	-	CR Maintained	Alive
2	84/M	1	III	181	1	2	2	De novo	CD10(+) GCB	40%	-	-	5	8C end	-	CR Maintained	Death Cholangio-carcinoma
3	84/F	2	IV	613	1	4	4	De novo	CD10(-) bcl-6(+) MUM-1(+) non-GCB	> 95%	-	+	2	8C end	-	CR Maintained	Alive
4	81/M	3	IV	3082	3 (BM)	5	5	De novo	CD10(+) GCB	10%	-	+	2	8C end	-	CR Maintained	Alive
5	83/M	1	IV	200	2 (BM)	2	3	De novo	CD10(+) GCB	90%	-	+	5	3C Discontinued	Self-discontinuation	NE	Alive
6	84/M	2	IV	194	3	4	4	De novo	CD10(+) GCB	90%	-	-	4	5C Discontinued	IP	CR Maintained	Death IP

Abbreviations; M, male; F, female; PS, performance status; LDH, lactate dehydrogenase; EN, extra nodal disease; BM, bone marrow; IPI, International Prognostic Index; CNS, central nervous system; COO, cell of origin; GCB, germinal center B-cell; FISH, fluorescence in situ hybridization; IHC, immunohistochemistry; -, negative or none; +, positive; CCI, Charlson comorbidity index; C, cycles; IP, interstitial pneumonia; CR, complete remission; NE, not examined

A swimmer plot illustrating the duration to last follow-up, treatment responses, timing of treatment discontinuation, and treatment outcomes for each patient is shown in Figure 1.

Fig. 1.

Swimmer plot showing the treatment duration, treatment response, and treatment outcomes for each patient

The median follow-up duration was 13.3 (3.3–34.3) months, and the median number of treatment cycles was 8 (3–8).

One patient discontinued outpatient follow-up at their own discretion. Complete remission (CR) was confirmed in the remaining five patients, including one patient in whom CR was confirmed at autopsy after death due to interstitial pneumonia (IP) (Figure 1, case 5). No cases of disease recurrence or CNS infiltration were observed.

Adverse events (AEs) observed during treatment are summarized in Table 2. These included grade 3 anemia in two patients, grade 3 thrombocytopenia in one patient, grade 4 leukopenia in one patient, grade 4 neutropenia in one patient, and grade 5 IP in one patient.

Table 2. Hematological and non-hematological AEs.

	n = 6
AE, n (%)	Any	Grade 3	Grade 4	Grade 5
HE	6 (100%)	3( 50%)	2 (33.3%)	0
Anemia	6 (100%)	2 (33.3%)	0	0
Thrombocytopenia	6 (100%)	1 (16.7%)	0	0
Leukopenia	4 (66.7%)	0	1 (16.7%)	0
Neutropenia	3 (50%)	0	1 (16.7%)	0
Non-HE	5 (83.3%)	0	1 (16.7%)	0
Na decreased	5 (83.3%)	0	0	0
K increased	4 (66.7%)	0	0	0
AST increased	3 (50%)	0	0	0
ALT increased	3 (50%)	0	0	0
Ca decreased	3 (50%)	0	0	0
Cr increased	2 (33.3%)	0	0	0
Stomatitis	2 (33.3%)	0	0	0
IP	1 (16.7%)	0	0	1 (16.7%)
Constipation	1 (16.7%)	0	0	0
Nausea	1 (16.7%)	0	0	0
Diarrhea	1 (16.7%)	0	0	0
AMY increased	1 (16.7%)	0	0	0

Abbreviations; AE, adverse events; HE, hematological events; Na, sodium; K, potassium; AST, aspartate aminotransferase; ALT, alanine aminotransferase; Ca,calcium; Cr, creatinine; IP, interstitial pneumonia; AMY, amylase

DISCUSSION

No patient in this study received CNS prophylaxis. Meta-analyses of retrospective clinical studies to date have shown that CNS prophylaxis is not effective in patients with DLBCL treated with either high-dose intravenous MTX or high-dose intrathecal MTX.¹⁵ Similar findings have been reported for CD5-positive DLBCL.¹ Therefore, CNS prophylaxis is not routinely administered to patients with DLBCL at our institution.

The mechanism underlying the poor prognosis of patients with CD5-positive DLBCL has not been fully elucidated; however, it has been attributed to the involvement of both CD5-mediated B-cell receptor (BCR)-dependent and BCR-independent pathways. The former involves inhibition of BCR signaling, whereas the latter is characterized by BCR-independent overexpression of interleukin-10, the antiapoptotic protein B-cell lymphoma 2, cyclin D2, and C-X-C motif chemokine receptor 4.³

CD5-positive DLBCL has been reported to harbor the MYD88 L265P mutation in 33%–52% of cases and the CD79B mutation in 38% of cases.^1,16,17 These genetic alterations are thought to contribute to the high frequency of CNS infiltration observed in this disease.^1,18 Patients with these mutations have been reported to have a poor prognosis and a suboptimal response to R-CHOP therapy.^16,19

Calabretta et al. used a recently described neural network–based probabilistic classifier, DLBclass, to assign each tumor to one of five genetic clusters (C1–C5). In their cohort of genetically analyzed patients, treatment arms and baseline clinical characteristics were well balanced. All 407 tumors were successfully assigned to a specific DLBclass cluster.

In patients with C5 DLBCL, which is characterized by recurrent MYD88 L265P and CD79B gain-of-function mutations and enhanced BCR signaling, the 5-year PFS was higher in the Pola-R-CHP arm than in the R-CHOP arm (70.4% [95% confidence interval (CI), 57.6–86.1] vs. 42.0% [95% CI, 28.0–63.0]). The hazard ratio for Pola-R-CHP versus R-CHOP was 0.38 (95% CI, 0.19–0.75; P = 0.005), suggesting that the polatuzumab-containing regimen mitigated the predicted poor outcome in this genetic subset.²⁰

Taken together, these results suggest that Pola-R-CHP therapy is effective in patients with CD5-positive DLBCL, who frequently harbor MYD88 L265P and CD79B mutations.

To date, only one study from China has evaluated the efficacy and safety of Pola-R-CHP therapy in CD5-positive DLBCL. In that study involving 27 patients, the overall response rate was 100% and the CR rate was 77.8%. The median follow-up period was 11.7 months, with PFS rates of 96.3% at 6 months and 85.2% at 12 months, suggesting favorable efficacy. However, the median age of patients in that cohort was substantially lower (64 [50–70] years), and patients aged ≥80 years were excluded.²¹ Consequently, the efficacy and safety of Pola-R-CHP therapy in CD5-positive DLBCL patients aged ≥80 years remained unclear.

In the present study, CR was achieved in five of six patients, and no cases of disease recurrence or CNS infiltration were observed, suggesting the effectiveness of this therapy (Table 1, Figure 1). All AEs observed during treatment, except for IP, were manageable (Table 2). Although the relationship between IP and CD5 positivity remains unclear, an association between IP and the Pola-R-mini-CHP regimen cannot be ruled out.²²

The efficacy of Pola-R-CHP therapy in patients with CD5-positive DLBCL who do not harbor MYD88 L265P and/or CD79B mutations remains unknown, and to date, no reports addressing this population have been published.

This study has some limitations. First, MYD88 L265P and CD79B mutational status was not assessed. Second, the sample size was small, and third, the observation period was relatively short. Consequently, the efficacy of the Pola-R-CHP regimen in the broader population of patients with CD5-positive DLBCL could not be definitively determined.

In conclusion, Pola-R-mini-CHP therapy may be effective in patients aged ≥80 years with CD5-positive DLBCL. Further accumulation of cases and longer-term follow-up are warranted to confirm these findings.

ETHICS

Ethics Committee Approval: This retrospective study was approved by the Ethics Committee of Saitama Cancer Center (Approval No. 1676; September 4, 2023).

INFORMED CONSENT

Informed consent was obtained from all participants included in the study.

AUTHOR CONTRIBUTIONS

Concept: K.A., Y.S., S.H., H.T., M.K., N.M., Y.I., M.K.; Design: K.K., Y.N., H.K., D.T., T.A., M.H., S.E., Y.K., T.N.; Data collection or processing: Y.S.; Analysis or interpretation: Y.S., H.K.; Writing: K.A., Y.S.

CONFLICT OF INTEREST

Y.S. reports receiving lecture fees from Chugai Pharmaceutical Co., Ltd., and AbbVie GK.

FINANCIAL DISCLOSURE

The authors declare that this study received no financial support.