Comparison of Existing Prognostic Models in Chronic Myelomonocytic Leukemia

Chronic myelomonocytic leukemia (CMML) is a rare blood cancer that straddles two worlds: myelodysplastic syndromes (MDS)—conditions where bone marrow makes abnormal blood cells—and myeloproliferative neoplasms (MPNs)—diseases where the body overproduces blood cells. Its rarity (fewer than 1 in 100,000 people are diagnosed annually) has left doctors without clear standards for predicting outcomes or choosing treatments. For patients, this uncertainty is daunting: Will the disease stay slow-growing, or will it turn into aggressive acute myeloid leukemia (AML)? How long can I expect to live?

To answer these questions, researchers from Lanzhou University set out to test six existing prognostic models for CMML—tools designed to rank patients by risk—to see which work best and where gaps remain. Their findings, published in the Chinese Medical Journal in 2020, offer valuable insights for managing this complex disease.

The Study: Who, What, and How

The team analyzed data from 45 CMML patients treated at the First Affiliated Hospital of Lanzhou University between 2008 and 2019. All patients provided written consent for their records to be used in research. Key tests included:

Blood work (white blood cells, hemoglobin, platelets, lactate dehydrogenase—LDH, a marker of cell damage).
Bone marrow exams (to check for abnormal cells and fibrosis).
Biopsies (for spleen or skin lesions) to confirm extramedullary infiltration (cancer spreading outside the bone marrow).

Statistical analysis linked these factors to two critical outcomes: overall survival (OS) (how long patients lived) and AML transformation (when CMML becomes more aggressive).

Key Findings: What Matters for Prognosis

The median age of patients was 65 (range: 18–83), and the median OS was 265 days—about 9 months. More than half (55.1%) died within a year, and 24.4% developed AML.

Risk Factors for Worse Outcomes

Four factors independently predicted shorter survival:

High white blood cell (WBC) counts: Elevated WBCs (a sign of uncontrolled cell growth) were strongly linked to worse outcomes (P = 0.001).
FAB subtype: A classification system for CMML; certain subtypes correlated with faster progression (P = 0.004).
High LDH: Elevated levels of this enzyme (released when cells are damaged) signal more aggressive disease (P = 0.013).
Abnormal cytogenetics: Changes in chromosomes (like trisomy 8 or complex abnormalities) worsened survival and AML risk (P = 0.002).

How Prognostic Models Performed

The team tested six models—CPSS, CPSS-MOL, MDAPS, G-MDAPS, GFM, and MMM—and found:

CPSS (CMML-Specific Prognostic Score) and CPSS-MOL (its molecular update) were most accurate at predicting AML transformation.
G-MDAPS (Global MD Anderson Prognostic Scoring System) best predicted death risk.
Extramedullary infiltration (spleen: 35.6% of patients; skin: 4.4%) did not strongly correlate with any model, though there was a weak link to the World Health Organization (WHO) CMML subtype (P = 0.028).

What Do These Models Actually Do?

Prognostic models are tools to help doctors tailor care. For CMML, they fall into two categories:

Traditional models: Use basic clinical data (age, blood counts, bone marrow blasts). Examples include MDAPS (from MD Anderson Cancer Center) and the original CPSS.
Molecular models: Add gene mutations—changes in DNA that drive cancer growth—to boost accuracy. Examples include CPSS-MOL, GFM, and MMM.

Here’s a quick breakdown of key models:

CPSS: Uses cytogenetics to predict survival and AML risk.
CPSS-MOL: Adds mutations in ASXL1, RUNX1, NRAS, and SETBP1—genes known to worsen outcomes—to CPSS.
GFM: A French model linking anemia, high WBC, low platelets, age >65, and ASXL1 mutations to shorter survival.
MMM: The Mayo Molecular Model refines an older tool by including ASXL1 mutations, which are present in 40–50% of CMML patients and strongly predict death.

These models help doctors choose treatments—like hypomethylating agents (e.g., decitabine) for high-risk patients or watchful waiting for low-risk ones—but they’re not perfect. For example, none clearly predict extramedullary infiltration, which can cause symptoms like splenomegaly (enlarged spleen) or skin lesions and may speed up AML transformation.

Extramedullary CMML: A Tricky Challenge

“Proliferative” CMML—where cancer spreads outside the bone marrow—can cause:

Splenomegaly: Enlarged spleen, common in CMML, linked to worse survival even with treatment.
Skin infiltration: Rare but serious; diagnosed via biopsy, it may signal AML is coming.
Meningeal involvement: Very rare, causing headaches, vision problems, or facial numbness. A lumbar puncture (spinal tap) can detect abnormal monocytes, which may disappear with remission.

While these cases are uncommon, they highlight a gap in current models: none are designed to predict or address spread beyond the bone marrow.

Limitations and Future Directions

The study had two key limitations:

Small sample size: Only 45 patients—understandable for a rare disease but limiting statistical power.
Retrospective design: Looking back at medical records can introduce bias (e.g., missing data).

Despite this, the research underscores two critical points:

Biopsies matter: For patients with extramedullary symptoms (e.g., skin lesions, enlarged spleen), biopsies are essential to confirm spread and guide treatment.
Better tools are needed: Existing models work for some outcomes but fail to predict extramedullary disease or rare complications. A more comprehensive tool—one that combines clinical data, cytogenetics, mutations, and extramedullary status—is urgently needed.

Final Takeaways

CMML is a rare, complex cancer with no cure. Prognostic models help doctors make informed decisions, but they’re not perfect. The Lanzhou University study reminds us that:

Key clinical factors (WBC, LDH, cytogenetics) still drive prognosis.
Molecular models (with gene mutations) are more accurate but need wider testing.
Extramedullary disease requires careful monitoring—even if it doesn’t fit neatly into current frameworks.

For patients, this means working closely with hematologists to track symptoms (like spleen enlargement or skin changes) and advocate for biopsies if needed. For researchers, it means prioritizing larger, prospective studies to build better prognostic tools.

This study was conducted by Jin-Li Jian, Yan-Hong Qiao, Shu-Ling Zhang, Hai-Zhen Ma, and Bei Liu from the First Clinical Medical College and Department of Hematology at the First Affiliated Hospital of Lanzhou University in China. It was published in the Chinese Medical Journal in 2020 under a CC-BY-NC-ND license, with all patients providing written consent.

doi: 10.1097/CM9.0000000000000637

Was this helpful?

0 / 0