Doxorubicin, bleomycin, vinblastine, and dacarbazine for Hodgkin lymphoma: Real-world experience from a Los Angeles County hospital

Abstract

Objective:

While there are significant ongoing advancements in the management of Hodgkin lymphoma, doxorubicin + bleomycin + vinblastine + dacarbazine remains a preferred option for early stage Hodgkin lymphoma and is a frequently used first-line treatment globally. The aim of this retrospective study is to analyze real-world doxorubicin + bleomycin + vinblastine + dacarbazine outcomes from a safety net hospital setting.

Methods:

This retrospective cohort consisted of 69 adult patients with classical Hodgkin lymphoma who received first-line doxorubicin + bleomycin + vinblastine + dacarbazine at Harbor-UCLA Medical Center from 2009 to 2024. Early (I–II) and advanced (III–IV) stage patients were included.

Results:

The median patient age was 41 years old (range 18–71). There was balanced distribution of early stage (7.2% stage I, 40.6% stage II) and advanced stage (20.3% stage III, 31.9% stage IV) Hodgkin lymphoma. With a median of six cycles (range 2–7) of doxorubicin + bleomycin + vinblastine + dacarbazine, the complete response rate was 78.3% and overall response rate was 82.6%. Five-year progression-free survival was 70.7% (70.2% for stage I–II, 71.3% for stage III–IV). Overall survival at 5 years was 95.4% (100% for stages I–II, 91.5% for stages III–IV). Bleomycin-associated lung toxicity was observed in 10 (14.5%) patients, including one treatment-related death.

Conclusions:

Response rates and overall survival from this real-world cohort are comparable to previously published contemporary studies. The high complete response rate, 5-year progression-free survival, and 5-year overall survival in this study further support the robust curative potential of doxorubicin + bleomycin + vinblastine + dacarbazine and validate its continued use in resource-limited settings.

Introduction

Globally, there is estimated 83,000 new cases of Hodgkin lymphoma (HL) diagnosed annually and 23,000 deaths, accounting for 0.4% and 0.2% of all cancer cases, respectively. ¹ In the United States, there are ~8500 new cases and 900 deaths annually. ² Mortality has gradually declined over the last 30 years, ³ and prognosis is generally favorable, with 5-year overall survival (OS) of 88.9%. ⁴

Advances in systemic therapy and radiation therapy (RT) have made HL one of the most curable cancers in both pediatric and adult populations, with a cure rate of roughly 80%. ⁵ Over the last 40 years, doxorubicin + bleomycin + vinblastine + dacarbazine (ABVD) has been a standard first-line treatment for classical HL. In the last decade, outcomes for patients with classical HL have improved with the incorporation of CD30-targeted antibody-drug conjugate brentuximab vedotin (BV)^{6 –8} and immune checkpoint inhibitors nivolumab^{9 –13} and pembrolizumab.^{10 –12} While BV-AVD ⁶ and nivolumab + AVD ¹³ have demonstrated superior efficacy in advanced stage (III–IV) HL, ABVD remains the preferred option for early-stage (I–II) disease^5,14,15 and a commonly used regimen globally.^16,17

Despite these advances, disparities remain among minority and low socioeconomic patients with HL. Population studies have shown that Hispanic^{18 –22} and Black^{18,20 –23} patients have worse outcomes compared to White patients. In addition, underserved status, including patients living in low socioeconomic neighborhoods,^22,24,25 as well as patients receiving income-based insurance such as Medicaid,^23,25,26 is associated with worse outcomes for HL patients in the United States. Specifically for ABVD, a 2021 retrospective study on a predominantly Hispanic population along the Texas-Mexican border showed a higher relapse rate compared to non-Hispanic White patients. ²⁷

Herein, we present a retrospective cohort of HL patients treated with first line ABVD from a Los Angeles County hospital and hypothesize that this regimen is effective and well tolerated in a safety net hospital setting.

Methods

Study design

A retrospective chart review was performed for all patients with HL at Harbor-UCLA Medical Center in Torrance, CA, USA from 2009 to 2024. Patients with nodular lymphocyte predominant HL were excluded. Other exclusionary criteria included first-line therapy other than ABVD, radiation alone, and no treatment (Figure 1). Patients with classical HL who received ABVD, ABVD with transition to AVD, or ABVD + RT were included. Data collection included demographic information, baseline clinical characteristics, treatment, response, adverse effects, progression-free survival (PFS), and OS. Written informed consent was waived per Institutional Review Board approval.

Figure 1.

Consort diagram depicting patient eligibility.

Staging

Patients were staged at the time of diagnosis according to the Ann Arbor staging system ²⁸ from 2009 to 2014 and the Lugano staging classification ²⁹ from 2015 to 2024. Presence or absence of B symptoms (fever, night sweats, unintentional weight loss >10%) was documented. Bulky disease was defined as any single tumor or nodal mass >10 cm or mediastinal mass ratio (MMR) >0.33 (MMR = width of mediastinal mass/intrathoracic diameter). Patients with stages I–II disease were classified as unfavorable risk if they had B symptoms, bulky disease, >3 nodal sites, or erythrocyte sedimentation rate >50 mm/h. Early stage patients with none of these unfavorable features were classified as favorable risk.

Therapy

Doxorubicin (25 mg/m²), bleomycin (10 U/m²), vinblastine (6 mg/m²), and dacarbazine (375 mg/m²) were administered intravenously on days 1 and 15 of a 28-day cycle for six cycles. Patients who received less than six cycles, due to response-adapted therapy or alternative reasons, were included. RT was used as part of first-line therapy based on provider discretion.

Response assessment

Response to therapy was assessed by positron emission tomography (PET), single-photon emission computed tomography (SPECT), or computed tomography (CT) using the International Working Group 2007 definitions ³⁰ from 2009 to 2014 and Lugano response criteria ²⁹ from 2015 to 2024. Imaging modality selection was dependent on date of therapy, provider selection, and logistical availability. Overall response rate (ORR) refers to patients with complete response (CR) or partial response (PR).

Toxicity monitoring and supportive care

Prior to chemotherapy initiation, assessment of ejection fraction was performed by echocardiography or multigated acquisition nuclear medicine scan. Baseline pulmonary function tests were performed and repeated in patients with lung disease or as clinically indicated for symptoms. Transfusions and growth factor support were utilized based on provider discretion. As a retrospective analysis, this study did not screen for adverse events utilizing a standardized protocol. Adverse event assessment was performed based on retrospective review of physician notes, laboratory data, and imaging studies with severity grading per Common Terminology Criteria for Adverse Events version 5.

Statistical analysis

Univariate analysis was used to compare demographics and baseline clinical characteristics between early and advanced stage patients. Kaplan–Meier curves were used for time-to-event analysis for OS and PFS, which was defined as time to disease progression, relapse, or death. An individual patient was censored at time of last follow-up. Log rank test was performed to assess for difference between curves. Hazard ratios were calculated with 95% confidence intervals (CIs). When assessing risk factors for bleomycin-associated lung toxicity, a two-sample T-test was used for age and z-score for two population proportions was used for human immunodeficiency virus (HIV) status and radiation. Significance level was defined by two-sided p < 0.05.

Results

Demographics

From 2009 to 2024, 69 patients with classical HL were treated with first line ABVD. The median age at diagnosis was 41 years old with an interquartile range (IQR) of 28–49. The cohort was diverse in terms of ethnicity/race (71% Hispanic, 13% Black, 6% White, 4% Asian, 6% other), and 6 (9%) patients were HIV positive. There was a relatively even distribution of early stage (47.8%) and advanced stage (52.2%) HL. There were no statistically significant differences in age, gender, ethnicity/race, B symptoms, HIV, or bulky disease between early and advanced stage patients (Table 1).

Table 1.

Demographics and baseline clinical information.

Demographics	All patients (n = 69)	Early stage (n = 33)	Advanced stage (n = 36)	p value
Age (years)
Median	41	35	44	0.230
Interquartile range	28–49	28–46	30–50
Range	18–71	23–71	18–63
Gender
Male	45 (65.2%)	22 (66.7%)	23 (63.9%)	0.809
Female	24 (34.8%)	11 (33.3%)	13 (36.1%)
Ethnicity/race
Hispanic	49 (71.0%)	27 (81.8%)	22 (61.1%)	0.058
Black	9 (13.0%)	2 (6.1%)	7 (19.4%)	0.099
White	4 (5.8%)	1 (3.0%)	3 (8.3%)	0.346
Asian	3 (4.3%)	0 (0%)	3 (8.3%)	0.090
Other	4 (5.8%)	3 (9.1%)	1 (2.8%)	0.262
Stage at diagnosis
I	5 (7.2%)
II	28 (40.6%)
III	14 (20.3%)
IV	22 (31.9%)
B symptoms	39 (56.5%)	22 (66.7%)	17 (47.2%)	0.104
HIV positive	6 (8.7%)	2 (6.1%)	4 (11.1%)	0.457
Bulky disease	13 (18.8%)	9 (27.3%)	4 (11.1%)	0.086

Doxorubicin + bleomycin + vinblastine + dacarbazine

Fifty (72.5%) patients received ABVD, and 19 (27.5%) patients received ABVD with transition to AVD with removal of bleomycin per the Response-Adapted Therapy in Advanced Hodgkin Lymphoma (RATHL) trial (n = 14) ³¹ or for pulmonary toxicity (n = 5). The median number of ABVD cycles was 6 (range 2–7), with 1 (1.4%) patient receiving seven cycles. Thirteen (18.8%) patients required dose reduction of at least one drug, including 10 (14.5%) patients for organ dysfunction (seven hepatic, three renal) and 3 (4.3%) patients for toxicity.

Response

Fifty-four (78.3%) patients had a CR to ABVD, 3 (4.3%) patients had a PR, 2 (2.9%) patients had stable disease, and 8 (11.6%) patients had progressive disease (Table 2). Response to ABVD was not evaluable for two patients (one death during cycle three from sepsis and multiorgan failure, one lost to follow-up). Response assessment was done by fluorodeoxyglucose (FDG) based PET in 45 patients, FDG–SPECT in 16 patients, and traditional CT in six patients.

Table 2.

Response to ABVD.

Response	All patients (n = 69)	Early stage (n = 33)	Advanced stage (n = 36)	Age <60 (n = 63)	Age >60 (n = 6)
CRR (± 95% CI)	78.3% ± 9.7%	72.7% ± 15.2%	83.3% ± 12.2%	76.2% ± 10.5%	100% ± 80.0%
ORR (± 95% CI)	82.6% ± 8.9%	78.8% ± 13.9%	86.1% ± 11.3%	81.0% ± 9.7%	100% ± 80.0%
Complete response	54 (78.3%)	24 (72.7%)	30 (83.3%)	48 (76.2%)	6 (100%)
Partial response	3 (4.3%)	2 (6.1%)	1 (2.8%)	3 (4.8%)	0
Stable disease	2 (2.9%)	1 (3.0%)	1 (2.8%)	2 (3.2%)	0
Progressive disease	8 (11.6%)	5 (15.2%)	3 (8.3%)	8 (12.7%)	0
Not evaluable	2 (2.9%)	1 (3.0%)	1 (2.8%)	2 (3.2%)	0

ABVD: doxorubicin + bleomycin + vinblastine + dacarbazine; CI: confidence interval; CRR: complete response rate; ORR: overall response rate.

Radiation

After completion of ABVD, 13 (18.8%) patients received RT with a median total radiation dose of 36 Gy across a median 20 fractions (Table 3). The most common indication was for bulky disease (n = 6), and the most common site was the mediastinum (n = 9).

Table 3.

Radiation therapy characteristics (n = 13).

Dose (Gy)	Median: 36 (range 12–40)
Fractions	Median: 20 (range 6–20)
Indication	Bulky disease (n = 6) H10U protocol (n = 2) Residual disease (n = 2) Provider selection (n = 2) High-risk features (n = 1)
Location	Six mediastinum Three mediastinum + supraclavicular Two axilla One cervical One submental

Progression-free survival

Six patients relapsed after initial CR (11.1% of CRs). A total of 19 (27.5%) patients had relapsed or refractory disease, all of whom received salvage systemic therapy. Ten patients received an autologous stem cell transplant, and two patients received an allogeneic transplant. With a median duration of follow-up of 71 months (IQR 32–109, range 1–186), 1- and 5-year PFS were 77.1% and 70.7%, respectively (Figure 2(a)). Five-year PFS was 81.8% for early stage favorable, 63.8% for early stage unfavorable, and 71.3% for advanced stage HL (Figure 2(b)). There were no significant differences in PFS for stages I–II favorable versus stages III–IV (HR 0.59, 95% CI 0.16–2.15, p = 0.480), stages I–II favorable versus stages I–II unfavorable (HR 0.50, 95% CI 0.13–1.91, p = 0.368), or stages I–II unfavorable versus stages III–IV (HR 1.15, CI 0.42–3.07, p = 0.773).

Figure 2.

Kaplan–Meier curves for (a) progression-free survival for all patients, (b) progression-free survival stratified by stage, (c) overall survival for all patients, and (d) overall survival stratified by stage. Tick marks on curves denote patients censored.

Overall survival

At the time of data analysis, seven patients were deceased including one death from bleomycin-induced interstitial pneumonitis, one death from sepsis and multiorgan failure during ABVD, four deaths from causes unrelated to HL while in remission, and one death from unknown cause while in remission. For the entire cohort, 5-year OS was 95.4% (Figure 2(c)). Patients with advanced stage HL had 5-year OS of 91.5%, compared to 100% for early stage favorable risk patients (Figure 2(d)). Log rank test showed no statistically significant differences in OS between stages I–II favorable, stages I–II unfavorable, and stages III–IV patients. Hazard ratios could not be calculated due to lack of events, including zero deaths in early stage patients.

Adverse events

Of the 69 patients in the study, 47 (68.1%) patients experienced at least one adverse event during treatment with ABVD (Table 4). Grade 3 or worse adverse events were observed in 24 (34.8%) patients with a rate of 31.7% in patients <60 years old and 66.7% in patients aged 60 or older (p = 0.086). The most common adverse events of any grade were neutropenia (43%), nausea (28%), and fatigue (25%). The most frequent grade 3 or worse events were neutropenia (32%), febrile neutropenia (3%), thrombocytopenia (3%), and pneumonitis (3%).

Table 4.

Adverse events.

Adverse events	Grade 1, n (%)	Grade 2, n (%)	Grade 3, n (%)	Grade 4, n (%)	Grade 5, n (%)	All grades, n (%)
Blood and lymphatic
Febrile neutropenia			2 (3)			2 (3)
Neutropenia	2 (3)	6 (9)	15 (22)	7 (10)		30 (43)
Thrombocytopenia	4 (6)		2 (3)			6 (9)
Gastrointestinal
Bloating	1 (1)	1 (1)				2 (3)
Constipation	4 (6)					4 (6)
Diarrhea	2 (3)					2 (3)
Nausea	14 (20)	5 (7)				19 (28)
Oral dysesthesia	1 (1)					1 (1)
Oral mucositis	2 (3)	2 (3)				4 (6)
Vomiting	5 (7)	2 (3)				7 (10)
General disorders and administration site conditions
Chills	1 (1)					1 (1)
Fatigue	15 (22)	2 (3)				17 (25)
Injection site reaction	3 (4)	1 (1)				4 (6)
Infections
Gallbladder infection			1 (1)			1 (1)
Lung infection		2 (3)				2 (3)
Sepsis					1 (1)	1 (1)
Skin infection		1 (1)				1 (1)
Upper respiratory infection		1 (1)				1 (1)
Urinary tract infection		2 (3)				2 (3)
Musculoskeletal
Bone pain	1 (1)					1 (1)
Myalgias	1 (1)					1 (1)
Nervous system
Headache	1 (1)					1 (1)
Peripheral motor neuropathy		1 (1)				1 (1)
Peripheral sensory neuropathy	4 (6)	1 (1)				5 (7)
Psychiatric
Anxiety	1 (1)	2 (3)				3 (4)
Depression		1 (1)				1 (1)
Renal and urinary
Dysuria	1 (1)					1 (1)
Hematuria		1 (1)				1 (1)
Respiratory and thoracic
Cough	1 (1)	1 (1)				2 (3)
Pneumonitis	3 (4)	2 (3)	1 (1)		1 (1)	7 (10)
Pulmonary fibrosis		2 (3)				2 (3)
Skin
Alopecia	1 (1)					1 (1)
Dry skin	1 (1)					1 (1)
Pruritus	2 (3)					2 (3)
Rash		1 (1)				1 (1)

Bleomycin-induced pulmonary toxicity was observed in 10 (14.5%) patients, including one treatment-related death. The median time from bleomycin initiation to lung toxicity was 4 months (range 2–6 months). For patients with bleomycin lung toxicity, median age was 49 years old, compared to 39 years of age for those without lung toxicity (p = 0.065). The rate of lung toxicity was 33.3% for HIV positive patients and 12.7% for HIV negative patients (p = 0.177). Pulmonary toxicity was observed in 23.1% of patients who received radiation and 12.5% of patients who did not get radiation (p = 0.327).

No cardiac toxicities were observed during therapy with ABVD. Three (4.3%) patients were diagnosed with congestive heart failure 5 years post-ABVD, including one patient who also had a myocardial infarction. None of the cardiac events were felt to be attributable to doxorubicin in light of preexisting cardiovascular disease, comorbidities, and other risk factors.

A total of four patients received granulocyte colony stimulating factor, including three for primary prophylaxis and one for secondary prophylaxis. No patients received erythropoiesis stimulating agents, thrombopoietin receptor mimetics, or prophylactic antimicrobials.

Second malignancies were observed in six patients, including two patients with second malignancies diagnosed concurrently with HL (Table 5). Among four patients diagnosed with a second malignancy after completion of HL therapy, the median time from ABVD initiation to diagnosis of second malignancy was 12.7 years (range 6.3–15.1 years).

Table 5.

Second malignancies.

Patient	HL treatment	Second malignancy	Time from ABVD to second malignancy (months)	Notes
Patient 1	ABVD ×6	Mantle cell lymphoma	75	Prostate cancer diagnosed concurrently with HL
Patient 2	ABVD ×2 + RT	LGL leukemia	126
Patient 3	ABVD ×6	Breast cancer	168	BRCA1 germline mutation
Patient 4	ABVD ×6	Carcinoma of unknown primary	181
Patient 5	ABVD ×6	Cutaneous squamous cell carcinoma	0	Deferred management of skin cancer
Patient 6	ABVD ×6	Renal cell carcinoma	0	Nephrectomy after ABVD

HL: Hodgkin lymphoma; RT: radiation therapy.

Discussion

This study presents the results of 15 years of HL patients treated with ABVD at a Los Angeles County safety net hospital with predominantly Hispanic and Black population. Five-year OS for patients with early stage (I–II) HL was 100%. This is comparable to previous studies of patients with stage IA or IIA nonbulky HL treated with ABVD, including a 2012 study with 12-year OS of 94% ¹⁴ and a 2016 Australian study with 5-year OS of 93%. ³² Five-year OS was 91.5% for advanced stage patients in our study, which is similar to previous studies of ABVD in advanced stage disease including a 2016 Australian study with 5-year OS of 89%, ³² a 2021 multinational (Italy, Israel, Spain) study with 30-month OS of 92.5%, ¹⁷ and a 2023 Thai study with 3-year OS of 94%. ³³ Across all stages, the 5-year OS was 95.4%, which aligns with Surveillance, Epidemiology, and End Results data. ⁴

The CR rate (CRR) was 78.3%, and the ORR was 82.6%. These response rates are broadly comparable with prior research.^14,31,32 Five-year PFS was 70.7% in our study. Previous ABVD studies report 5-year PFS rates ranging from 62% to 87%.^{14,17,32 –34}

Compared to prior studies on minority populations in the United States, our study reports favorable outcomes. In a 2016 population study, at a mean follow-up time of 11 years, Black (80.6%) patients show decreased OS than compared to non-Hispanic White (87.0%) and Hispanic (86.8%) patients. ²⁵ A 2022 study on patients aged 15–39 enrolled in the Medicaid Program in New York showed statistically significant differences in 5-year OS between Black (87%), Hispanic (90%), and non-Hispanic White (94%) patients. ²³ Further studies involving Hispanic populations demonstrate numerically worse outcomes; however, the results did not achieve statistical significance.^35,36

Bleomycin-induced lung injury is a potentially life-threatening adverse effect, affecting 5%–10% of HL patients overall and 10%–20% of adults >60 years old treated with ABVD. ³⁷ Mortality from bleomycin pulmonary toxicity ranges between 0.2% and 8.4%. ²⁶ In our study, bleomycin lung toxicity was present in 14.5% of patients, including one treatment-related death, and all cases were diagnosed 2–6 months from bleomycin initiation. The notable incidence of and mortality from lung toxicity observed reinforces the potential risks of bleomycin. There were signals for potential risk factors including older age and HIV, however without reaching statistical significance. Evidence is mixed regarding OS with bleomycin omission.^{38 –40} With the significant morbidity associated with bleomycin and the emergence of novel therapies, current practice guidelines recommend limited bleomycin cycles, and for advanced stage disease, complete omission of bleomycin. Novel approaches with liposomal doxorubicin without bleomycin for elderly patients are under ongoing investigation. ⁴¹

No cardiac events were observed during treatment with ABVD, and the three cases of cardiomyopathy observed 5 years post-ABVD were not thought to be related to anthracycline toxicity. Nevertheless, more robust and structured monitoring, work-up, and management of cardiotoxicity in patients with HL could be beneficial and may be associated with improved ejection fraction. ⁴²

This study is limited in a few ways. First, the sample size of 69 patients is relatively small compared to other studies, many of which included more than 100 patients.^17,20,23 In addition, this is a retrospective study, and there is potential for confounding patient factors. However, it must be recognized that many contemporary ABVD studies are retrospective studies,^{20,23 –25} as the treatment regimen has been well-established for some time. The study did not incorporate a prespecified sample size calculation or justification, which may represent an additional limitation of the study. Another consideration when interpreting the data are lack of standardization of the number of ABVD cycles, including one patient who received seven cycles. Lastly, patient follow-up was inconsistent, thereby limiting survival analysis beyond 5 years.

The significance of this study is that it offers a unique perspective on a primarily underserved patient population. Harbor-UCLA, as a Los Angeles County safety net hospital, sees many patients who are uninsured or underinsured. Previous research from our group has described suboptimal response rates and survival for Harbor-UCLA patients with acute myeloid leukemia, ⁴³ as well as decreased OS for HL patients overall. ⁴⁴ Potential contributing factors include complex medical comorbidities, decreased health literacy, social barriers such as housing insecurity, lack of transportation, financial hardship, and lack of caregiver support which constrain compliance, and decreased access to transplant, novel therapies, and clinical trials.

Of note, this study presents outcomes comparable to those of previous studies, the majority of which were done in more resource-rich settings. This may be a result of the mature status of the ABVD regimen at this time with robust toxicity management, response-adapted therapy, incorporation of radiation, and improvements in salvage therapy and transplant. This study highlights the importance of characterizing real-world outcomes, particularly from resource-limited settings. In resource-limited settings globally without access to BV or nivolumab, ABVD can be a highly effective first-line treatment with high cure rates.

At the present, ABVD remains a standard of care for stages I–II HL patients. Early stage HL patients in our study demonstrate 100% 5-year OS with ABVD. In the case of early stage unfavorable disease, novel regimens have been studied achieving comparable outcomes while mitigating toxicities using nivolumab + AVD ⁴⁵ or better results than ABVD using BV-AVD. ⁷ For advanced stage HL, ABVD remains useful under limited circumstances. The standard of care, as well as future research directions, have shifted towards the use of nivolumab + AVD, ¹³ BV-AVD, ⁶ and BV, etoposide, cyclophosphamide, doxorubicin, dacarbazine, and dexamethasone ⁴⁶ with other novel combinations under ongoing investigation. Future efforts to promote equitable access to these therapies and real-world corroboration of clinical trial data are imperative.

Conclusion

In this single-center retrospective study analyzing ABVD in HL, the CRR was 78.3%. Five-year PFS and OS were 70.7% and 94.5%, respectively. Adverse effects, including a significant rate of lung toxicity, were comparable to previous research. The impressive response rates and survival in an underserved and diverse patient population reinforce the effectiveness of ABVD in resource-constrained settings, while potential risks of bleomycin toxicity must also be considered.