Carmustine

證據等級: L5

預測適應症: 10 個

Carmustine: From Malignant Glioma to Lymph Node Cancer

One-Sentence Summary

Carmustine (BCNU) is a nitrosourea-class alkylating agent originally established for malignant glioma treatment — including as Gliadel® biodegradable implant wafers (FDA-approved 1997) and as the backbone of high-dose conditioning chemotherapy for stem cell transplantation. The TxGNN model predicts it may be effective for Lymph Node Cancer (encompassing lymphoma and related lymphoid malignancies), with 7 clinical trials and 20 publications currently supporting this direction.

Quick Overview

Item	Content
Original Indication	Malignant glioma (FDA-approved as Gliadel® implant; not registered in Singapore)
Predicted New Indication	Lymph Node Cancer
TxGNN Prediction Score	98.49%
Evidence Level	L2
Singapore Market Status	Not Marketed
Number of Registrations	0
Recommended Decision	Proceed with Guardrails

Why is This Prediction Reasonable?

Carmustine (BCNU, 1,3-bis(2-chloroethyl)-1-nitrosourea) works by alkylating DNA — specifically forming O6-chloroethyl adducts at guanine residues that spontaneously rearrange into interstrand cross-links. These cross-links block DNA replication and transcription, triggering apoptosis in actively dividing tumour cells. Detailed MOA data were not available in the Singapore regulatory database (carmustine is not registered locally); however, the mechanism is extensively documented in published oncology literature. Critically, carmustine is highly lipid-soluble, allowing broad tissue penetration — including lymphoid tissue.

The connection between carmustine and lymph node cancer is both mechanistically grounded and clinically practice-established. Lymphoma cells — particularly diffuse large B-cell lymphoma (DLBCL) and Hodgkin lymphoma — frequently express low levels of the DNA repair enzyme MGMT (O6-methylguanine DNA methyltransferase). Because MGMT repairs exactly the type of adduct carmustine creates, low MGMT expression renders these tumour cells intrinsically sensitive to carmustine's lethal cross-linking. This is the same molecular vulnerability exploited in glioma treatment, making the biological leap from brain tumour to lymphoma mechanistically coherent. Carmustine is the "B" (BCNU) in the internationally used BEAM conditioning regimen (Carmustine + Etoposide + Cytarabine + Melphalan) administered before autologous stem cell transplantation (ASCT) in relapsed/refractory lymphoma.

Multiple completed Phase 2 trials support this application, including the largest lymphoma PBSC transplant registry (NCT00345865, n=473) and a real-world R-BEAM practice study (NCT01702961, n=75). While the only Phase 3 trial (NCT00002772) was terminated early and did not target lymphoma directly, BEAM has been recognized internationally as the standard myeloablative conditioning regimen for relapsed/refractory lymphoma ASCT since the 1990s. The TxGNN score of 98.49% reflects the strong knowledge-graph signal between carmustine and lymphoid malignancy — consistent with decades of clinical practice.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00345865	Phase 2	Completed	473	Largest lymphoma PBSC transplant registry; directly evaluates carmustine-based conditioning; eight different high-dose regimens compared, providing the strongest real-world evidence base for this indication
NCT01702961	N/A	Completed	75	R-BEAM (rituximab + BEAM) current-practice study in lymphoma/Hodgkin's disease after first relapse or incomplete initial response; confirms real-world use pattern of carmustine in ASCT preconditioning
NCT01141712	Phase 2	Completed	43	BEAM conditioning regimen in HIV-infected patients with aggressive B-cell lymphoma and Hodgkin lymphoma; directly tests carmustine's role in ASCT for lymphoid malignancies in a challenging patient subgroup
NCT00002772	Phase 3	Terminated	602	Phase 3 randomized comparison of intensive carmustine-containing high-dose vs. lower-dose consolidation for breast cancer with 4–9 involved axillary lymph nodes; terminated early — incomplete efficacy data, limiting conclusions
NCT01008462	Phase 2	Completed	16	Sequential autologous + haploidentical allogeneic HCT for high-risk Hodgkin lymphoma, non-Hodgkin lymphoma, multiple myeloma, and CLL using carmustine-containing conditioning; demonstrates feasibility
NCT01468311	Phase 1/2	Terminated	6	Y-90-daclizumab radioimmunotherapy + BEAM for refractory Hodgkin lymphoma; terminated early with only 6 participants — data insufficient, but concept supports carmustine's role in HL

Literature Evidence

PMID	Year	Type	Journal	Key Findings
18988286	2008	RCT	Cancer	5-year randomized Phase 2 trial (n=158) in high-risk Hodgkin lymphoma: intensive chemotherapy vs. ABVD followed by myeloablative ASCT; carmustine-containing myeloablative conditioning directly tested; establishes carmustine's role in HL consolidation
7577204	1995	RCT	J Natl Cancer Inst Monogr	CALGB 9082: Randomized comparison of high-dose vs. low-dose cyclophosphamide+cisplatin+carmustine with ABMT for breast cancer patients with ≥10 involved axillary lymph nodes; provides Phase 3-level evidence for carmustine activity in lymph node-positive malignancy
15767638	2005	RCT	J Clin Oncol	CALGB 9082/SWOG 9114/NCIC MA-13 multi-group prospective randomized comparison; high-dose carmustine-based chemotherapy with stem cell support vs. intermediate-dose for high-risk breast cancer with multiple axillary nodes
2642573	1989	Case Series	Leukemia	23 heavily pre-treated Hodgkin's disease patients received BCNU 450–600 mg/m² + etoposide + cyclophosphamide + ABMT; first systematic clinical evidence of carmustine's activity in lymphoid malignancy with ASCT support
10473086	1999	Cohort	Clin Cancer Res	O6-alkylguanine-DNA alkyltransferase (MGMT) expression in cutaneous T-cell lymphoma (mycosis fungoides); low MGMT correlates with carmustine sensitivity — mechanistic support for the MGMT-pathway rationale in lymphoma
15505610	2004	Cohort	Biol Blood Marrow Transplant	5-year outcomes of high-dose cyclophosphamide+carmustine+thiotepa (CBT) + AHST for high-risk primary breast cancer with ≥10 positive axillary lymph nodes; RFS and prognostic factor analysis
11895898	2002	Cohort	Clin Cancer Res	PK/PD analysis of high-dose carmustine in 85 patients with breast cancer and ≥10 involved lymph nodes; identifies pharmacokinetic predictors of survival and toxicity, supporting dose optimization
36213836	2022	Cohort	J Oncology	Retrospective review of high-dose carmustine-containing chemotherapy + ASCT in locally advanced triple-negative breast cancer; marginal survival benefit in receptor-defined subgroup when standard chemotherapy fails
11063378	2000	Cohort	Biol Blood Marrow Transplant	1,111 consecutive breast cancer patients at 5 California centres receiving carmustine-based high-dose chemotherapy; treatment-related mortality 2.3%; hematopoietic graft source — not stage — determines TRM
1639635	1992	Case Series	Int J Radiat Oncol Biol Phys	49 women with breast cancer involving ≥10 axillary nodes received high-dose cyclophosphamide+cisplatin+carmustine (HDCT) + ABMT followed by post-mastectomy radiotherapy (CALGB); safety and feasibility established

Cytotoxicity

Item	Content
Cytotoxicity Classification	Conventional cytotoxic — Nitrosourea class (bifunctional alkylating agent)
Myelosuppression Risk	High — hallmark delayed-onset myelosuppression; neutrophil and platelet nadirs typically occur 4–6 weeks after administration and may persist for an additional 1–2 weeks; cumulative toxicity risk increases with repeated cycles
Emetogenicity Classification	Moderate to High (particularly at the high doses used in ASCT conditioning)
Monitoring Items	CBC with differential (at least weekly for 6 weeks post-dose), liver function tests (ALT/AST/bilirubin), renal function (creatinine), pulmonary function tests (critical: cumulative pulmonary fibrosis is dose-limiting; monitor closely when lifetime cumulative dose approaches 1,400 mg/m²)
Handling Protection	Must follow cytotoxic drug handling regulations; gloves, gown, and eye protection required; carmustine is a skin and mucous membrane irritant — avoid contact; diluted IV solution is stable for limited time and must be protected from light

Safety Considerations

Carmustine is not registered with the Singapore Health Sciences Authority (HSA); no local regulatory safety data are available.

Please refer to the FDA or EMA-approved prescribing information for complete safety details. Key known risks include: delayed cumulative myelosuppression (onset 4–6 weeks post-dose, duration up to 6 weeks — the most common dose-limiting toxicity); pulmonary toxicity and fibrosis (insidious onset, dose-limiting at cumulative doses exceeding 1,400 mg/m²; may be fatal); hepatotoxicity (elevated transaminases and veno-occlusive disease reported); ocular irritation (transient conjunctival flushing and visual blurring during IV infusion); and secondary malignancies (therapy-related myeloid neoplasms with prolonged alkylator exposure). Drug-drug interaction data were not available in this Evidence Pack — a formal DDI review using an established clinical pharmacy database is required before prescribing.

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: Carmustine is the internationally accepted backbone of the BEAM myeloablative conditioning regimen for autologous stem cell transplantation in relapsed/refractory lymphoma — an application supported by decades of clinical practice, multiple completed Phase 2 trials (including a registry study with n=473), and consistent mechanistic evidence linking MGMT-low lymphoma cells to carmustine sensitivity. The TxGNN score of 98.49% is fully concordant with this established clinical knowledge base. The primary barrier to Singapore use is the absence of local registration, not insufficient evidence.

To proceed, the following is needed:

Establish a regulatory pathway with HSA for importation or compassionate use of carmustine (IV formulation and/or Gliadel wafer); carmustine has no Singapore registration
Obtain the FDA or EMA-approved package insert for complete warnings, contraindications, and dosing guidance before any clinical use
Conduct a formal drug-drug interaction assessment using a validated clinical pharmacy database (DDI data were absent from this Evidence Pack)
Clarify the intended clinical context: systemic BEAM conditioning for lymphoma ASCT vs. local Gliadel wafer delivery (mechanistically and logistically distinct applications requiring separate protocols)
Verify institutional ASCT programme capacity and haematology/oncology support infrastructure if BEAM conditioning is the intended route
Supplement with DrugBank MOA data to complete the mechanistic section of any institutional review submission
Consider MGMT promoter methylation status as a prospective biomarker for patient selection, given its established predictive role for nitrosourea sensitivity
Disclaimer

This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.