Waldenstrom Macroglobulinemia

Waldenstrom Macroglobulinemia: Mechanistic Summary

2026-04-13
Manual MONDO:0100280 Model: n/a 7 citations

Waldenstrom Macroglobulinemia: Mechanistic Summary

Disease anchor and modeling choice

Waldenstrom macroglobulinemia (WM) is best modeled here as a single disease-level mechanism graph rooted to MONDO:0100280 Waldenstrom macroglobulinemia, not as a lattice of every oncology subclass. The key defining combination is:

  • lymphoplasmacytic lymphoma involving the bone marrow
  • secretion of a monoclonal IgM paraprotein
  • recurrent MYD88-driven B-cell oncogenesis with CXCR4-mutant subclonal evolution

Following issue #1198, I treated lymphoplasmacytic lymphoma as the disease parent / diagnostic histopathology concept rather than opening a separate dismech page for every related ontology refinement. Likewise, I did not split smoldering WM, familial WM, refractory WM, recurrent WM, or transformed WM into separate disorder files, because those are disease-state, predisposition, or transformation facets rather than distinct causal programs for the baseline WM mechanism graph. In the YAML, this was reflected by keeping one disease page and encoding subtype structure only as flat has_subtypes facets for clinical course and molecular context.

Relevant ontology anchors:

  • MONDO:0100280 Waldenstrom macroglobulinemia
  • MONDO:0000432 lymphoplasmacytic lymphoma
  • NCIT:C80307 Waldenstrom Macroglobulinemia
  • NCIT:C3212 Lymphoplasmacytic Lymphoma

Foundational disease definition

PMID:12720118 remains the key disease-definition paper for this curation. It defines WM as an uncommon lymphoproliferative disorder characterized by bone marrow lymphoplasmacytic infiltration and IgM monoclonal gammopathy, and it explicitly argues that WM is a distinct clinicopathologic entity rather than just an IgM secretion syndrome.

PMID:11718214 complements this by describing the underlying cell composition: small mature B lymphocytes, plasmacytoid lymphocytes, and plasma cells. For the dismech entry, that supports treating the bone-marrow lymphoplasmacytic clone as the core disease unit and the IgM paraprotein as a downstream biochemical output of that clone.

Core mechanisms selected for the YAML graph

1. Bone marrow lymphoplasmacytic infiltration

This is the disease-defining tissue state. The mechanistic unit is the marrow infiltrating clone composed of B-cell/plasma-cell lineage lymphoplasmacytic cells. This node should sit near the top of the graph because it links the founding genomic lesions to the clinical consequences of marrow occupation and paraprotein secretion.

Key support:

2. Monoclonal IgM secretion

The IgM paraprotein is the key biochemical output of the malignant clone and is best modeled as a separate downstream node rather than bundled into the marrow infiltration node. This helps keep the graph atomic and allows direct edges to paraprotein-driven complications.

Key support:

3. Serum hyperviscosity

Hyperviscosity is a downstream biophysical consequence of heavy circulating IgM burden. I modeled this as a separate pathophysiology node rather than collapsing it into the phenotype list, because it is an intermediate pathogenic process that explains neurologic and visual symptoms.

Key support:

4. MYD88 L265P founder mutation

MYD88 L265P is the dominant recurrent somatic lesion in WM and should be treated as an upstream genomic driver node. The cached abstract set supports two important, conservative claims:

  • MYD88 L265P is present in the majority of WM cases (PMID:24224040)
  • CXCR4 WHIM-like mutations are usually acquired after MYD88 L265P in WM oncogenesis (PMID:26659815)

Because the cached abstract set was used conservatively, I did not overstate the full MYD88 -> BTK -> IRAK -> NF-kB signaling chain in the YAML evidence block without a validated PMID cache line for that exact claim. That mechanistic detail is real and well-supported in the broader WM literature, but the YAML entry is restricted to what could be quoted exactly and validated in this branch.

5. CXCR4 WHIM-like subclonal evolution

CXCR4 mutations are not the founding lesion; they are modeled as a later, subclonal branch layered onto the MYD88-mutant clone. This is a good example of an oncology refinement that belongs inside the same disease graph rather than in a separate disease page.

Key support:

6. Mast-cell supportive signaling in the marrow microenvironment

PMID:18216294 shows that excess marrow mast cells are common in WM and provide growth/survival signals to lymphoplasmacytic cells. This is worth a separate node because it captures a bona fide microenvironmental mechanism rather than a mere descriptive pathology feature.

Key support:

Clinical phenotype and complication summary

The best abstract-backed phenotype set from the cached reference slice is:

Hyperviscosity-related neurologic disorders are common enough to matter mechanistically but were better represented as a downstream pathophysiology node than as an HP-grounded phenotype because the available ontology grounding in this schema is much cleaner for the manifestations than for the syndrome label itself.

Treatment conclusions for the entry

The current disease-level treatment section should emphasize:

  • watchful waiting for asymptomatic patients, per the 2023 IWWM consensus (PMID:37099027)
  • fixed-duration chemoimmunotherapy as an important first-line option, especially bendamustine-rituximab (PMID:37099027)
  • zanubrutinib as a current covalent BTK inhibitor option with lower toxicity and deeper remissions than ibrutinib in the referenced randomized-trial update (PMID:37099027)
  • plasmapheresis for rapid paraprotein reduction in hyperviscosity and related neurologic complications (PMID:18813229)

Following #1198, NCIT should be preferred where it gives materially better oncology specificity. In the YAML this means:

  • NCIT:C3212 for the histopathology finding
  • NCIT:C1702 for rituximab
  • NCIT:C141428 for zanubrutinib
  • NCIT:C15304 for plasmapheresis

I initially attempted to ground BR and zanubrutinib with NCIT regimen terms as well, but the current RegimenTerm validator expansion in this branch did not accept those codes, so I retained the validated disease-level treatment representation using treatment actions plus ontology-grounded agents/procedures.

References