The Miller Lab The Miller Lab The Miller Lab
  • Our Approach
  • Meet The Team
  • Publications
  • Presentations
  • Posts
  • Software
  • Funding
  • Slides
  • Donate

The Prognostic Value of the Merkel Cell Polyomavirus Serum Antibody Test: A Dual Institutional Observational Study

David M. Miller

Sophia Z. Shalhout

Kayla M. Wright

Matt A. Miller

Howard L. Kaufman

Kevin S. Emerick

Harrison T. Reeder

Ann W. Silk

Manisha Thakuria

2024-05-02

Original Research · Cancer

The Prognostic Value of the Merkel Cell Polyomavirus Serum Antibody Test

A dual-institutional study of whether baseline AMERK serostatus and antibody titer can refine prognosis and risk stratification in Merkel cell carcinoma.

Merkel Cell Carcinoma AMERK Merkel Cell Polyomavirus Prognostic Biomarker Risk Stratification
Journal Cancer
Published online May 2, 2024
Final citation 2024 · Volume 130 · Issue 15

Citation

Miller DM, Shalhout SZ, Wright KM, Miller MA, Kaufman HL, Emerick KS, Reeder HT, Silk AW, Thakuria M. The prognostic value of the Merkel cell polyomavirus serum antibody test: A dual institutional observational study. Cancer. 2024;130(15):2670–2682. doi:10.1002/cncr.35314.

View DOI View on PubMed Read Full Article View Preregistration

Why this study matters

Merkel cell carcinoma can behave very differently among patients who otherwise appear similar by conventional staging. AMERK testing detects circulating antibodies against Merkel cell polyomavirus oncoproteins and is already used clinically for surveillance in patients who are seropositive.

We asked a related but distinct question: does the initial AMERK result also contain prognostic information at diagnosis?

Central Finding

Baseline AMERK seropositivity was associated with more favorable outcomes, but the prognostic signal was concentrated in patients with localized disease. The initial antibody titer also tracked closely with measures of tumor burden.

The cohort at a glance

Total MCC cohort 531

Patients identified

Patients treated at two academic health systems from 2015 through 2022.

AMERK tested 362

Tested during care

Sixty-eight percent of the overall MCC cohort underwent AMERK testing.

Primary cohort 261

Tested within 90 days

These patients formed the primary analysis cohort for baseline prognostic assessment.

Seropositive 49.4%

129 of 261

Approximately half of patients had an initial positive AMERK result.

What we found

  • Initial AMERK seropositivity was associated with a lower risk of recurrence after adjustment for clinical covariates.
  • The same direction of association was observed for event-free, overall, and MCC-specific survival.
  • Initial antibody titer was strongly associated with clinical stage, tumor stage, nodal stage, tumor size, and overall disease extent.
  • The prognostic association of seropositivity was concentrated in patients presenting with localized disease.
  • Among patients with stage III or IV disease, outcomes were more similar regardless of baseline serostatus.
The Stage-Specific Signal

The most clinically interesting finding may be in stage I and II disease. Within a group conventionally labeled “localized,” baseline AMERK serostatus appeared to identify patients with meaningfully different outcome profiles.

What does the initial titer tell us?

Serostatus may reflect biology

A positive test identifies patients mounting a detectable antibody response to viral oncoproteins, potentially capturing biologic information not represented by anatomic stage alone.

Titer may reflect burden

Among seropositive patients, higher initial titers were associated with greater disease burden across several complementary measures.

Baseline prognosis versus surveillance

These are related but distinct uses of the same biomarker. The baseline AMERK test may help establish prognostic context at diagnosis. Serial measurements in a seropositive patient can then be used as a longitudinal surveillance tool.

Seronegative patients do not have that longitudinal antibody marker available, which may increase the importance of clinical examination, imaging, and complementary blood-based approaches.

A Useful Distinction

The first AMERK result helps define who the patient is biologically and prognostically. Subsequent AMERK measurements ask a different question: whether disease burden may be changing over time.

How we approached the analysis

  • The statistical analysis plan was preregistered before analysis.
  • Competing-risk methods were used for recurrence and MCC-specific mortality.
  • Models incorporated clinically relevant covariates including age, sex, stage, immune status, performance status, and initial treatment.
  • Planned sensitivity analyses examined borderline AMERK values, systemic therapy, missing data, and alternative endpoint definitions.
  • Multiplicity and proportional-hazards assumptions were evaluated explicitly.

What this study does not establish

  • The study was retrospective and observational; serostatus should not be interpreted as causing better outcomes.
  • The analyses were exploratory rather than built around a single prespecified significance threshold.
  • Overall and MCC-specific survival analyses contained relatively few events and therefore greater uncertainty.
  • AMERK is not a universal biomarker because only approximately half of patients are seropositive at diagnosis.
  • The findings support refinement of risk assessment, not replacement of conventional staging or clinical judgment.
The Practical View

An AMERK test obtained near diagnosis can provide more than a future surveillance baseline. In localized MCC, it may add clinically useful prognostic information that complements anatomic stage and helps frame how intensively an individual patient should be followed.

Authors

David M. Miller, Sophia Z. Shalhout, Kayla M. Wright, Matt A. Miller, Howard L. Kaufman, Kevin S. Emerick, Harrison T. Reeder, Ann W. Silk, and Manisha Thakuria.

This page summarizes selected findings from the published article and is not intended to reproduce the complete manuscript. See the journal publication for the complete methods, effect estimates, figures, tables, sensitivity analyses, disclosures, and supporting information. This site represents our opinions only. See our full Disclaimer and Terms of Use Agreement.