CurrentColorado Rocky Mountain Health PlansPolicy 2026T0588OO

Molecular Oncology Testing for Solid Tumor Cancer Diagnosis, Prognosis, and Treatment Decisions

Policy governing coverage of non-FDA-cleared molecular and genomic tests used for diagnosis, prognosis, and treatment decisions in solid tumor cancers for UnitedHealthcare Commercial and Individual Exchange plans.

Policy Summary

PayerColorado Rocky Mountain Health Plans

PolicyMolecular Oncology Testing for Solid Tumor Cancer Diagnosis, Prognosis, and Treatment Decisions

Policy CodePolicy 2026T0588OO

Change TypeNo material clinical changes

Effective DateMay 1, 2026

Next Review DateN/A

Key ActionProvide documentation that the test indication and specimen meet policy criteria and that results will be used to influence the member's management decision.

SourceLink

6 testsBreast GEP listed as proven

≤50 genesPanels considered proven when ≤

>50 genesPanels generally unproven when >

6,693MINDACT participants

10,004Oncotype/RS meta-analysis participants

Trek Health ingests and normalizes Transparency in Coverage data and payer policy updates to give provider organizations a clear view of how commercial reimbursement behaves across markets, payers, and services. Our platform transforms raw payer disclosures into structured intelligence that supports contract evaluation, payer negotiations, and service line strategy. By combining market benchmarks with ongoing policy visibility, Trek helps teams identify variability, risk, and opportunity in commercial reimbursement. The result is faster insight, stronger negotiating positions, and more informed financial decisions.

Coverage Criteria — Molecular & Genomic Testing

Breast Cancer Gene Expression Profiling (GEP)

Proven and medically necessary when ALL of the following criteria are met for newly diagnosed invasive breast cancer:

Patient and disease timing: New diagnosis within the last 6 months

Nodal status: Lymph node negative (including micrometastases ≤2 mm) OR 1–3 positive ipsilateral axillary lymph nodes

Metastatic status: No distant metastases

Receptor status: Hormone receptor–positive (ER and/or PR positive) AND HER2 receptor–negative

Treatment intent and feasibility: Adjuvant chemotherapy is not precluded by other factors (e.g., advanced age or significant comorbidities)

Applies to MammaPrint, Oncotype DX Breast, Prosigna (PAM50), Breast Cancer Index (BCI), and EndoPredict

During adjuvant hormonal therapy (alternate covered scenario): Currently receiving adjuvant hormonal therapy and testing is intended to guide decision on extended adjuvant hormonal therapy after discussion between patient and treating physician

Applies to listed GEP tests

Lung Cancer Molecular Profiling

Covered when used for non-small cell lung cancer and consistent with limits:

Tumor type: Diagnosis of non-small cell lung cancer (adenocarcinoma or nonsquamous histology, or other NSCLC when clinical features indicate higher probability of a driver)

Panel size limits: Multigene molecular profiling panels of up to 50 genes are proven and medically necessary; panels >50 genes are permitted only when used consistent with FDA-cleared/approved companion diagnostic policy<= 50 genes

Applies to tissue or liquid biopsy (cfDNA/ctDNA)

Specimen considerations: Testing may be performed on tissue or, if tissue is inadequate, by validated plasma cfDNA/ctDNA methods per guideline recommendations

Use cfDNA when tissue insufficient; validated PD-L1 assays used per guidelines

Prostate Cancer Gene Expression Profiling (GEP)

Proven and medically necessary when ALL specified clinical conditions are met:

Applicable tests and setting: Biopsy-proven, untreated, localized adenocarcinoma of the prostate with no clinical evidence of metastasis or lymph node involvement; test ordered by a physician specializing in organ-confined prostate cancer (urology, radiation oncology, or medical oncology)

Applies to Genomic Prostate Score (GPS/Oncotype DX Prostate), Prolaris, Decipher Prostate Biopsy genomic classifier

Clinical candidacy for active surveillance/definitive therapy: Results will be used to assist treatment decision‑making when the individual is a candidate for either active surveillance or definitive therapy; life expectancy >10 yearslife expectancy > 10 years

Risk group specified below

Risk group indications: For GPS: very low, low, or favorable intermediate risk. For Prolaris/Decipher biopsy tests: very low, low, favorable intermediate, unfavorable intermediate, or high risk

Thyroid Molecular Testing

Covered when ALL of the following are met:

Cytology category: Fine needle aspiration cytology is indeterminate for follicular-pattern pathology (Bethesda III or IV)

Impact on management: Test results will be used to make decisions about further surgery (e.g., proceed to diagnostic lobectomy vs observation/repeat FNA)

Applies to Afirma GSC, ThyroSeq v3, ThyGeNEXT/ThyraMIR

Anaplastic thyroid cancer: Comprehensive genomic profiling (CGP) is proven and medically necessary for confirmed anaplastic thyroid cancer

CGP for ATC

Multiple tests limitation: Use of more than one molecular profile test for the same indeterminate thyroid nodule is unproven and not medically necessary

Uveal Melanoma GEP

Covered when ALL of the following are met:

Disease and setting: Primary, localized uveal melanoma; no evidence of metastatic disease

Test naïveté: Individual has not previously had DecisionDx-UM testing for the current diagnosis

DecisionDx-UM

Management linkage: Prognostic stratification results will be used to influence follow-up intensity or treatment planning

Document how result will change monitoring or management

Unproven / Not Medically Necessary Molecular Tests

The following are considered unproven and not medically necessary (examples):

Large NGS/CGP panels not otherwise specified: NGS panels >50 genes or comprehensive genomic profiling (CGP) unless explicitly specified as covered> 50 genes

Generally not medically necessary per policy

MRD and multi-cancer detection: Measurable residual disease (MRD) assays in solid tumors (tumor-informed or tumor‑naïve) and multi‑cancer early detection (MCED/MCD) blood‑based tests

Examples: Signatera, Guardant Reveal, Galleri, other MRD/MCED tests

Named tests with insufficient evidence: Specific assays judged unproven/not medically necessary due to insufficient validated clinical utility, including but not limited to: CancerTYPE ID, Decipher Bladder, DecisionDx-Melanoma, DiffDx-Melanoma, DecisionDx-SCC, DermTech PLA, Merlin, myPath Melanoma, ExoDx Prostate, MyProstateScore, Confirm mdx, Select mdx, PancraGEN, NavDx (pending evidence) and whole exome/genome/transcriptome/methylation profiling for routine use

Assay-specific clinical intent and evidence

Clinical evidence and intended applications summarized for gene-expression assays in breast cancer

Oncotype DX — intent and evidence: Oncotype DX (21-gene) provides a Recurrence Score prognostic for distant recurrence and can inform adjuvant chemotherapy decisions in ER+/HER2- early breast cancer including node-negative and select node-positive (1–3 nodes) patients; randomized trials and meta-analyses support prognostic and selective predictive utility

RS categories and trial-specific cutpoints vary by study

Prosigna (PAM50) — intent and evidence: Prosigna (PAM50, 50 genes) provides a ten-year distant recurrence risk estimate and intrinsic subtype in postmenopausal women with HR+ N0 or N1 disease and is not intended for N2 diseaseNot for N2 disease

Prognostic in specified populations

MammaPrint/BluePrint — intent and evidence: MammaPrint (70-gene) classifies tumors as genomic low- or high-risk and, with BluePrint, may influence systemic therapy recommendations; MINDACT and prospective observational studies demonstrate impact on chemotherapy decisions in selected populations

Evidence summaries (no coverage rules provided)

Evidence summaries — informational: This section contains study results, trial follow-up, and observational data (e.g., Oncotype DX RCTs, MINDACT, EndoPredict trials, BCI analyses) summarizing prognostic/predictive associations but does not itself establish additional coverage rules

Informational only

ASCO-based coverage criteria

Covered when ALL of the following guideline-supported conditions are met

ASCO-recommended population: Postmenopausal individuals or those >50 years with early-stage ER+, HER2- breast cancer who are node-negative or have 1–3 positive nodes may be tested with Oncotype DX, MammaPrint, BCI, or EndoPredict to guide adjuvant endocrine and chemotherapy decisionsage>50 or postmenopausal

ASCO 2022 guidance

Assays for node-negative postmenopausal: BCI and Prosigna may be used in postmenopausal individuals with node-negative ER+, HER2- breast cancernode-negative

ASCO 2022

Premenopausal guidance: For premenopausal individuals, Oncotype DX may be used in node-negative ER+, HER2- breast cancer per guideline specificspremenopausal + node-negative

Assay-specific evidence (informational)

Assay-specific evidence and utility observations (informational only)

MammaPrint/BluePrint observations: MammaPrint/BluePrint testing has changed systemic therapy recommendations in prospective studies (e.g., ~29% change in WSG‑PRIMe) and MINDACT demonstrates selective omission of chemotherapy in some clinical‑high/genomic‑low patients with small absolute differences in DMFS

Informational: physician adherence high in studies

EndoPredict observations: EndoPredict (EPclin) has shown changes in adjuvant chemotherapy recommendations (≈36% change in a prospective trial) and prospective data on predictive utility exist though limitations remain

Informational only

BCI and extended endocrine therapy: BCI provides prognostic information and pooled analyses suggest predictive value for benefit from extended endocrine therapy though independent assessments note limitations and need for further high-quality evidence

Informational only

Guideline-based indications for genomic assays

Guideline-based recommendations for use of genomic assays:

ASCO 2022 summary: Oncotype DX, MammaPrint, BCI, and EndoPredict may be used to guide adjuvant endocrine and chemotherapy in postmenopausal or >50 patients with ER+, HER2- early BC who are node-negative or have 1–3 positive nodes; Oncotype DX may be used in premenopausal node‑negative patients; BCI may be used to consider extended endocrine therapy after five years

ASCO 2022

NCCN stance: NCCN prefers the 21‑gene Oncotype DX for prognosis and prediction of chemotherapy benefit and recognizes other assays provide prognostic information though may not predict chemo benefit; GEPs complement but do not replace staging

NCCN v6.2024

NICE 2024 notes: NICE allows EndoPredict, Oncotype DX or Prosigna to guide chemo decisions in postmenopausal (and selected male/trans/non-binary/intersex profiles) ER/PR+, HER2- early BC with 1–3 nodes but not for premenopausal women with 1–3 nodes; MammaPrint not recommended for some node‑positive premenopausal contexts

NICE 2024

Oncotype DX DCIS

Oncotype DX DCIS evidence summary:

Evidence quality for DCIS score: Published evidence for Oncotype DX DCIS is low quality; limited data suggest association with radiotherapy use but clinical outcomes from score‑guided decisions are unknown

Hayes assessment 2024

NSCLC molecular testing and liquid biopsy

Molecular testing in NSCLC — evidence and guideline recommendations:

Use of multiplexed tissue NGS: Multiplexed NGS panels or broad hybrid-capture assays are supported to detect actionable genomic alterations in NSCLC and are preferred over multiple single-gene tests where available

Pathology/oncology society guidance; tissue preferred

Liquid biopsy role: Plasma cfDNA/ctDNA (liquid biopsy) may be used when tissue is limited or insufficient, but not as a routine replacement for tissue testing; serial ctDNA changes may have prognostic value though heterogeneity exists

ASCO and CAP/AMP guidance; meta-analysis supports prognostic association

Panel size and success considerations: Panel success rates vary by specimen type and quality; optimize specimen quantity/quality to maximize detection of driver alterations

Specimen adequacy affects success rates (e.g., Oncomine study)

Coverage-relevant clinical criteria and evidence context

Coverage-relevant evidence summaries and contexts where testing may be useful:

Prognostic association in prostate cancer: Higher scores from genomic classifiers (GPS, Decipher, Prolaris) are associated with increased risk of biochemical recurrence, metastasis, and prostate cancer–specific mortality, though evidence quality varies

Meta-analyses and cohort studies summarized

Potential clinical utility settings: Genomic testing may inform (1) active surveillance versus definitive therapy selection, (2) post‑prostatectomy adjuvant therapy decisions, and (3) risk stratification to guide intensification or de‑intensification of therapy when results would change management

Selective use recommended by societies

Evidence limitations: Evidence that genomic-test–guided management improves long‑term clinical outcomes is limited or immature; many studies are retrospective or of low quality and manufacturer involvement is common

Hayes and other assessments note limitations

Selective use for tissue-based genomic biomarkers

Covered when ALL of the following are met (per professional guidance applicability):

Commercial availability and intended impact: Test is commercially available (e.g., Oncotype DX Prostate/GPS, Prolaris, Decipher, ProMark) AND the assay result, when considered with routine clinical factors, is likely to affect management decisions

ASCO/AUA/ASTRO guidance: selective (not routine) use

When testing may be considered

Coverage consideration when genomic/molecular testing may be appropriate

Prostate genomic testing — conditional use: Commercial tests (Decipher, GPS/Oncotype DX Prostate, Prolaris) may be offered when results combined with routine clinical factors are likely to change management; routine ordering is not recommended

Decipher may inform post‑RP adjuvant therapy (NCCN category 2B); ensure management impact documented

Active surveillance/biopsy decision context: In newly diagnosed men eligible for active surveillance, MRI and genomics may be used selectively when results would alter management; urine/serum adjunct markers may inform biopsy decisions

AUA/SUO/NCCN conditional recommendations

Thyroid nodule molecular testing considerations

Molecular/genomic testing for indeterminate thyroid nodules (Bethesda III/IV) — evidence summaries

Rule-out performance: Afirma GSC and ThyroSeq v3 demonstrate high sensitivity and negative predictive value in meta-analyses and prospective cohorts and may help avoid diagnostic surgery in many benign nodules, though specificity and PPV vary and reference‑standard limitations existsensitivity ~95% in pooled analyses

Document local malignancy prevalence and test selection

Expanded assays and limitations: Expanded expression/fusion detection panels (e.g., Afirma XA) and larger variant panels identify alterations of potential therapeutic implication but clinical utility for altering outcomes is not established

Analytical validation exists but outcome data limited

Considered/Conditional Uses

Contextual coverage considerations and guideline-based scenarios where molecular testing may be considered

Thyroid indeterminate cytology: Molecular testing (mutation panels, miRNA classifiers, multigene GCs) may supplement malignancy risk assessment for AUS/FLUS or FN/SFN nodules and influence management when results would change the choice between observation, repeat FNA, or diagnostic surgery

Selection based on clinical question and local performance characteristics; counsel patients on limitations

Cutaneous melanoma — SLNB decision: 31‑GEP/i31‑GEP may inform sentinel lymph node biopsy decisions in selected T1–T2 tumors and has been associated with reduced SLNB rates in prospective studies but clinical outcome impact remains uncertain

Use cautiously; AAD/NCCN advise prospective validation and reserved clinical application

Adjunct diagnostic testing for equivocal melanocytic lesions: Ancillary molecular techniques (CGH, FISH, GEP) may be used as adjuncts to expert dermatopathology for diagnostically equivocal lesions

Interpret within clinical/pathologic context

Coverage-relevant evidence synthesis

Summary of policy-relevant findings and guideline positions from this section:

Melanoma prognostic testing stance: Prognostic GEP and other molecular prognostic tests for cutaneous melanoma have shown associations with outcomes but have not demonstrated added clinical utility beyond clinicopathologic factors; routine prognostic molecular testing is discouraged until better use criteria are defined

AAD and NCCN caution; prospective validation needed

Uveal melanoma prognostic testing: When biopsy is performed for uveal melanoma, molecular or chromosomal prognostic testing (including GEP) may be used to stratify metastatic risk and guide follow-up intensity, but evidence of impact on outcomes is limited

NCCN allows prognostic stratification if biopsy performed; DecisionDx‑UM evidence limited

cSCC 40‑GEP consideration: 40‑GEP (DecisionDx‑SCC) can stratify metastatic risk (Class 1/2A/2B) and may inform adjuvant radiation decisions in high‑risk cSCC when multidisciplinary rationale is documented; evidence has limitations including potential manufacturer involvement

Use with multidisciplinary justification

Coverage rationale and criteria

Coverage recommendations and rationale summarized from guideline statements and evidence reviews

CUP testing approach: Consider molecular profiling (GEP or NGS/CGP) for CUP only after comprehensive clinicopathologic evaluation and histologic workup; use judiciously given limited demonstrated benefit for GEP‑directed site‑specific therapy

Prefer tissue testing; consider cfDNA if tissue not feasible

When DNA/NGS favored in CUP: DNA mutational profiling/CGP may provide added diagnostic value and identify actionable alterations for targeted therapy or clinical trial enrollment; some studies show DNA may resolve more CUPs than GEP

Use to identify actionable variants and trial eligibility

cfDNA blood tests for CRC screening: cfDNA blood‑based screening tests show promising sensitivity for colorectal cancer but low sensitivity for advanced precancerous lesions; further longitudinal outcome data are required before routine screening use

ECLIPSE study findings; further evidence needed

Guideline summary (CRC)

Guideline-based recommendations summarized for colorectal and tumor‑profiling guidance

ACG on SEPT9: American College of Gastroenterology recommends against SEPT9 (mSEPT9) testing for CRC screening due to low sensitivity and limited evidence

ACG 2021 guideline

ASCO on tumor profiling for adjuvant CRC: ASCO notes tumor‑based profiling tools are not yet ready for routine use to inform adjuvant therapy decisions; additional evidence required

ASCO guidance

NCCN on CRC biomarker testing: NCCN recommends universal MMR/MSI testing for colorectal cancer and testing for KRAS/NRAS/BRAF/HER2 in advanced disease; multigene panels, Immunoscore, and post‑surgical ctDNA currently lack sufficient data for routine adjuvant decision‑making

NCCN Colon Cancer v5.2024

Evidence summary (Pancreatic)

Pancreatic cyst/tumor molecular testing evidence

Insufficient evidence for routine pancreatic molecular risk assessment: Overall evidence is insufficient to support routine molecular testing for pancreatic cancer risk assessment or diagnosis; high‑quality studies are needed

Hayes assessment and guideline summaries

PancreaSeq/targeted NGS utility: Targeted NGS panels (eg, PancreaSeq) on cyst fluid show high sensitivity and specificity for identifying mucinous cysts and advanced neoplasia in prospective cohorts and improve when combined with cytopathology

Studies report sensitivities ~88–93% and specificities ~95–100% in surgical cohorts

TTNB NGS feasibility and risks: EUS‑guided through‑the‑needle biopsy (TTNB) combined with NGS can achieve high diagnostic accuracy for mucinous cysts/IPMN but carries procedural adverse event risk (~9.9%) and study limitations

Single‑center and cohort limitations noted

Summary of evidence-based uses and limitations

Evidence-supported uses and limitations summarized from the excerpt

TTNB/NGS diagnostic value: NGS on TTNB or pancreatic cyst fluid can identify KRAS/GNAS and other mutations to aid diagnosis of mucinous cysts and IPMNs and detect markers of advanced neoplasia (TP53/PIK3CA/PTEN); sensitivities and specificities reported are high but studies have limitations

Consider procedural risk and case selection

cfDNA methylation and MCED panels: cfDNA methylation assays and pan‑GI/MCED panels show promising diagnostic accuracy in preliminary studies but lack prospective population‑level utility data for screening

Further prospective evaluation required

NavDx (TTMV‑HPV DNA): TTMV‑HPV DNA assays demonstrate high specificity and moderate‑to‑high sensitivity for diagnosing and surveilling HPV+ OPSCC in cohort studies, but evidence on clinical utility and impact on outcomes is limited

Hayes and ECRI assessments note limited evidence; correlate results clinically

Evidence-based coverage considerations

Summary of coverage-relevant findings and limitations

Signatera/ctDNA MRD stance: Current evidence is insufficient to support routine use of tumor‑informed or tumor‑naïve MRD ctDNA assays (eg, Signatera) for monitoring response or surveillance across solid tumors; prospective trials pending

Hayes and NICE conclude insufficient evidence

NavDx diagnostic performance: NavDx reports high specificity (~100%) and sensitivity (~91.5% diagnostic, ~88.4% surveillance) in cohorts but false negatives occur and clinical utility to change outcomes is unproven

Ferrandino et al.; Hayes assessment

Actionability limitation for ctDNA/TTMV results: Detection of plasma ctDNA or TTMV‑HPV DNA without concurrent clinical, radiographic, or pathologic correlates lacks actionable therapeutic implications outside clinical trials

NCCN and guideline statements recommend clinical correlation

The policy lists multiple named tests and broad categories as unproven and not medically necessary. Examples expressly included are NGS panels >50 genes or comprehensive genomic profiling (CGP), multi-cancer early detection (MCED/MCD) tests (e.g., Galleri), measurable residual disease (MRD) assays in solid tumors (tumor‑informed or tumor‑naïve such as Signatera, Guardant Reveal, RaDaR), selected melanoma and prostate diagnostic/triage assays (DecisionDx‑Melanoma, DermTech PLA, ExoDx Prostate, MyProstateScore, Confirm mdx, Select mdx), NavDx, PancraGEN and PancreaSeq, and whole exome/genome/transcriptome or whole‑genome methylation profiling. These tests are identified as examples and are considered unsupported by the available evidence unless specifically addressed elsewhere in the policy.

The document includes extensive applicable procedure and PLA/Category III codes for molecular and genomic oncology assays. Examples shown in the policy include PLA/Category III and other proprietary codes (e.g., 0011M, 0012M, 0013M, 0016M, 0020M, 0005U, 0018U, 0019U), cfDNA and targeted NGS panel codes (e.g., 0487U, 0498U, 0538U), and higher‑level CPT panel codes (e.g., 81445, 81455–81464) as well as specific oncology multigene expression assay CPT/HCPCS entries (e.g., 81518–81523, 81540, 81542, 81552). These code listings and their descriptions are provided for reference and reflect the range of test technologies and panel sizes represented in the policy.

The Prosigna (PAM50) assay is described as a 50‑gene FFPE tumor GEP intended to provide a ten‑year distant recurrence risk and intrinsic subtype for postmenopausal women with HR+, N0 or N+ (up to N1) disease. Importantly, Prosigna is not intended for individuals with N2 disease (four or more positive lymph nodes) and should not be used for that population.

The policy reiterates that the Prosigna (PAM50) assay is not intended for individuals with N2 disease (four or more positive lymph nodes), consistent with the assay's intended use description.

Gene expression profiling assays beyond the named, guideline‑supported tests are not supported when evidence is lacking. The policy specifically notes that other breast GEPs and related assays (for example, Theros H/I, DCISionRT, certain 41‑gene or 76‑gene signatures and other platforms not otherwise described) currently lack sufficient evidence to demonstrate clinical utility and therefore are not supported for routine clinical use.

Per professional guidance summarized in the policy, genomic assays are not recommended for individuals with HER2‑positive or triple‑negative breast cancer. Additionally, gene expression assays do not replace standard histopathologic staging and biomarker assessment; they are adjuncts to clinical and pathologic information when used in recommended, guideline‑specified populations.

Independent evidence reviews (Hayes assessments) identified insufficient or very low‑quality evidence to support routine use of certain assays in specific prostate cancer populations. The assessments conclude that available data do not robustly demonstrate improved long‑term clinical outcomes from routine ordering of these genomic tests in some groups, and therefore the evidence is considered insufficient for routine coverage in those contexts.

The policy cautions against offering additional molecular biomarkers that lack adequate performance data or are not commercially validated. Examples include emerging panels and tests (e.g., Select mdx, Confirm mdx) that have been judged by evidence reviews to have limited or very low‑quality supporting data; such biomarkers should not be routinely offered until sufficient peer‑reviewed evidence demonstrates clinical validity and utility.

Guideline statements captured in the policy emphasize that additional molecular biomarkers without sufficient supporting data should not be offered routinely. Professional societies recommend selective use of commercially available molecular tests only when results, together with routine clinical factors, are likely to change management; routine ordering is discouraged.

The policy identifies tests that lack demonstrated clinical utility or are supported only by very low‑quality evidence (for example, some prognostic GEPs, pigmented lesion assays, and certain adjunct diagnostic tests). Independent HTA and evidence assessments (Hayes, Ontario HTA, systematic reviews) reported limited, inconsistent, or biased evidence and concluded that clinical utility has not been established for many of these assays; consequently they are considered unsupported pending higher‑quality outcome data.

Covered Indications — When Testing May Be Covered

Not Covered / Unproven Tests

Policy summary statements identify categories considered not covered unless otherwise specified: NGS panels >50 genes or comprehensive genomic profiling (CGP), measurable residual disease (MRD) assays in solid tumors (tumor‑informed or tumor‑naïve such as Signatera), multi‑cancer early detection (MCED/MCD) tests, and routine whole exome/genome/transcriptome sequencing. The policy lists multiple named tests and platform classes as examples of modalities that are not supported for routine coverage in the absence of clear evidence demonstrating clinical utility.

The policy includes code listings and descriptive code text as reference material; this excerpt provides the code descriptors but does not itself contain explicit not‑covered language beyond the code catalog.

Hayes assessments indicate there is insufficient evidence to support use of Oncotype DX for predicting chemotherapy benefit in individuals with N2 disease (four to nine positive lymph nodes). The policy therefore does not support using Oncotype DX to guide adjuvant chemotherapy in patients with N2 nodal status.

Clinical trial data cited in the policy indicate that Prosigna (PAM50) provides prognostic information but was not shown to be predictive of radiotherapy benefit in subgroup analyses; Prosigna's intended use remains prognostication in specified HR+, postmenopausal, N0/N1 populations rather than predicting radiotherapy responsiveness.

Evidence regarding BluePrint used with MammaPrint is described as limited. Studies have shown molecular subtyping and traditional pathological subtyping may disagree in some cases, and the combined use of BluePrint with MammaPrint currently has insufficient evidence to fully support clinical utility in all contexts; caution is advised when interpreting results that conflict with pathology.

Applicable Codes and Coding Guidance

Applicable CPT/PLA/other codes referencedmixed

0011M	Oncology, prostate cancer, mRNA expression assay of 12 genes (10 content and 2 housekeeping), RT-PCR test utilizing blood plasma and urine, algorithms to predict high-grade prostate cancer risk.
0012M	Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes (MDK, HOXA13, CDC2 [CDK1], IGFBP5, and CXCR2), utilizing urine, algorithm reported as a risk score for having urothelial carcinoma.
0013M	Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes, utilizing urine, algorithm reported as a risk score for having recurrent urothelial carcinoma.
0016M	Oncology (bladder), mRNA, microarray gene expression profiling of 219 genes, utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as molecular subtype.
0020M	Oncology (central nervous system), analysis of 30000 DNA methylation loci by methylation array, utilizing DNA extracted from tumor tissue, diagnostic algorithm reported as probability of matching a reference tumor subclass.
0005U	Oncology (prostate) gene expression profile by real-time RT-PCR of 3 genes (ERG, PCA3, and SPDEF), urine, algorithm reported as risk score.
0018U	Oncology (thyroid), microRNA profiling by RT-PCR of 10 microRNA sequences, utilizing fine needle aspirate, algorithm reported as a positive or negative result for moderate to high risk of malignancy.
0019U	Oncology, RNA, gene expression by whole transcriptome sequencing, formalin-fixed paraffin embedded tissue or fresh frozen tissue, predictive algorithm reported as potential targets for therapeutic agents.
0022U	Targeted genomic sequence analysis panel, non-small cell lung neoplasia, DNA and RNA analysis, 23 genes, interrogation for sequence variants and rearrangements.
0026U	Oncology (thyroid), DNA and mRNA of 112 genes, next-generation sequencing, fine needle aspirate of thyroid nodule, algorithmic analysis reported as a categorical result.

1–10 of 13

1/2

Applicable molecular/genomic oncology codes (PLA/Category III/U and CPT-level panels)mixed

0011M	Oncology, prostate cancer, mRNA expression assay of 12 genes (10 content and 2 housekeeping), RT-PCR test utilizing blood plasma and urine, algorithms to predict high-grade prostate cancer risk.
0012M	Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes (MDK, HOXA13, CDC2 [CDK1], IGFBP5, and CXCR2), utilizing urine, algorithm reported as a risk score for having urothelial carcinoma.
0013M	Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes, utilizing urine, algorithm reported as a risk score for having recurrent urothelial carcinoma.
0016M	Oncology (bladder), mRNA, microarray gene expression profiling of 219 genes, utilizing FFPE tissue, algorithm reported as molecular subtype.
0020M	Oncology (central nervous system), analysis of 30000 DNA methylation loci by methylation array, utilizing DNA extracted from tumor tissue, diagnostic algorithm reported as probability of matching a reference tumor subclass.
0005U	Oncology (prostate) gene expression profile by real-time RT-PCR of 3 genes (ERG, PCA3, and SPDEF), urine, algorithm reported as risk score.
0018U	Oncology (thyroid), microRNA profiling by RT-PCR of 10 microRNA sequences, utilizing fine needle aspirate, algorithm reported as a positive or negative result for moderate to high risk of malignancy.
0019U	Oncology, RNA, gene expression by whole transcriptome sequencing, FFPE or fresh frozen tissue, predictive algorithm reported as potential targets for therapeutic agents.
0022U	Targeted genomic sequence analysis panel, non-small cell lung neoplasia, DNA and RNA analysis, 23 genes, interrogation for sequence variants and rearrangements, reported as presence or absence of variants and associated therapy(ies) to consider.
0026U	Oncology (thyroid), DNA and mRNA of 112 genes, next-generation sequencing, fine needle aspirate of thyroid nodule, algorithmic analysis reported as a categorical result.

1–10 of 99

1/10

Applicable CPT/HCPCS Codesmixed

81462	Solid organ neoplasm, genomic sequence analysis panel, cell-free nucleic acid (e.g., plasma), interrogation for sequence variants; DNA analysis or combined DNA and RNA analysis, copy number variants and rearrangements.
81463	Solid organ neoplasm, genomic sequence analysis panel, cell-free nucleic acid (e.g., plasma), interrogation for sequence variants; DNA analysis, copy number variants, and microsatellite instability.
81464	Solid organ neoplasm, genomic sequence analysis panel, cell-free nucleic acid (e.g., plasma), interrogation for sequence variants; DNA analysis or combined DNA and RNA analysis, copy number variants, microsatellite instability, tumor mutation burden, and rearrangements.
81479	Unlisted molecular pathology procedure.
81504	Oncology (tissue of origin), microarray gene expression profiling of > 2000 genes, utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as tissue similarity scores.
81518	Oncology (breast), mRNA, gene expression profiling by real-time RT-PCR of 11 genes (7 content and 4 housekeeping), utilizing formalin-fixed paraffin-embedded tissue, algorithms reported as percentage risk for metastatic recurrence and likelihood of benefit from extended endocrine therapy.
81519	Oncology (breast), mRNA, gene expression profiling by real-time RT-PCR of 21 genes, utilizing formalin-fixed paraffin embedded tissue, algorithm reported as recurrence score.
81520	Oncology (breast), mRNA gene expression profiling by hybrid capture of 58 genes (50 content and 8 housekeeping), utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as a.
81521	Oncology (breast), mRNA, microarray gene expression profiling of 70 content genes and 465 housekeeping genes, utilizing fresh frozen or formalin-fixed paraffin-embedded tissue, algorithm reported as index related to risk of distant metastasis.
81522	Oncology (breast), mRNA, gene expression profiling by RT-PCR of 12 genes (8 content and 4 housekeeping), utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as recurrence risk score.

1–10 of 22

1/3

Covered CPT CodesCPT

No codes listed

Panel size threshold — policy summary

Panel size thresholdPanels greater than 50 genes are generally considered unproven and not medically necessary unless specifically identified as covered in the policy

Exception for companion diagnosticsPanels >50 genes may be allowed when used consistent with FDA‑cleared/approved companion diagnostic policy (e.g., as noted for NSCLC)

Coding implicationUse CPT/PLA codes that reflect panel size (e.g., 81445 for 5–50 gene panels; 81455/81456 for 51+ gene panels) when documenting test scope

Panel gene count ranges — examples for coding

Small to intermediate panels (example)5–50 genes — represented by CPT descriptor 81445/81449 (solid organ neoplasm panels specifying 5–50 genes)

Provider Actions, Prior Authorization & Documentation

Prior Authorization

Plan Applicability & Prior Authorization

Prior authorization may be required for molecular oncology and genomic tests depending on the member-specific benefit plan. Providers should verify plan applicability (Commercial and Individual Exchange) and obtain prior authorization when indicated by the benefit document.

Applicable to UnitedHealthcare Commercial and Individual Exchange plans — check member benefit for prior authorization requirements.
Medical records documentation may be requested to confirm clinical criteria for coverage.

Billing Rule

Codes Relevant to Prior Authorization / Workflow

Use the listed CPT/HCPCS/PLA codes that correspond to the exact test performed. Billing with codes not listed or with a mismatched code and test may trigger claim denial.

Key Definitions & Test Descriptions

Comparative Genome Hybridization (CGH)

DefinitionComparative Genome Hybridization (CGH) is a technology that detects genomic copy number variations by hybridizing labeled tumor and reference DNA on arrays and measuring probe intensity differences

OutputReports indicate gains or losses relative to reference DNA and can provide copy‑number and gene‑differentiating information

Use caseUsed as an adjunct diagnostic technique for resolving equivocal melanocytic lesions and other tumor classification contexts

Comprehensive Genomic Profiling (CGP)

DefinitionComprehensive Genomic Profiling (CGP) is an NGS-based test able to detect all classes of genomic alterations (SNVs, indels, CNAs, rearrangements, MSI, TMB) from a single specimen

Typical specimen and scope

Eligibility Requirements (Excerpt Notes)

These excerpts did not contain top‑level eligibility requirement nodes in the provided segments.

Background & Evidence Context

Comprehensive genomic profiling (CGP) and gene expression profiling (GEP) are molecular approaches used to detect somatic genomic alterations or tumor mRNA expression signatures to guide diagnosis, prognosis, and treatment decisions. Liquid biopsy (plasma cfDNA/ctDNA) is a noninvasive method to detect tumor‑derived nucleic acids for molecular characterization or monitoring. These technologies have distinct intended uses and specimen requirements and should be used in the clinical contexts supported by evidence and professional guidelines.

Coding Guidance — Key Metrics & Thresholds

Panel size threshold — coding reminder

Reminder — coding guidancePanels >50 genes are generally considered unproven/not medically necessary unless specifically included as covered (policy exception), so coders should flag large panel requests for prior review

Action for codersVerify panel gene count against CPT/PLA descriptors (e.g., 81445 for 5–50 genes; 81455/81456 for 51+ genes; large U‑codes for specific PLA panels) when assigning codes

Prior authorization implicationTests exceeding 50 genes may require documentation of clinical indication and adherence to companion diagnostic policy to be considered

Panel gene count ranges — coder guidance

Common ranges and examples5–50 genes (CPT 81445/81449) ; 51+ genes (CPT 81455/81456) ; very large panels reported in PLA/U codes (e.g., 521‑gene 0585U, 600+ gene 0538U)

Revision History

2026-05-01policy_effectiveLatest

Policy effective date published (Medical Policy 2026T0588OO).

2026-05-01codes_update

Applicable CPT/PLA/other procedure codes list updated; added code 0630U and archived prior policy version 2026T0588NN.

Policy Summary

PayerColorado Rocky Mountain Health Plans

PolicyMolecular Oncology Testing for Solid Tumor Cancer Diagnosis, Prognosis, and Treatment Decisions

Policy CodePolicy 2026T0588OO

Change TypeNo material clinical changes

Effective DateMay 1, 2026

Next Review DateN/A

Key ActionProvide documentation that the test indication and specimen meet policy criteria and that results will be used to influence the member's management decision.

SourceLink

On This Page

Assay-specific risk group applicability

Post-prostatectomy (Decipher Prostate RP): Decipher Prostate RP genomic classifier is proven to inform adjuvant treatment after radical prostatectomy when adverse pathological features are present (e.g., Gleason ≥8, extracapsular extension, positive margins, seminal vesicle invasion) OR for PSA > 0 after prostatectomy

Applies to Decipher Prostate RP

Ordering restrictions and unsupported tests: Prostate molecular screening panel tests (e.g., ExoDx Prostate, MyProstateScore, Confirm mdx, Select mdx) are considered unproven and not medically necessary

Molecular screening panels not covered

See policy exclusions list for specifics

Evidence includes MINDACT and WSG-PRIMe

Not recommended subtypes: No gene expression assays are recommended for individuals with HER2-positive or triple‑negative breast cancerHER2+ or triple-negative

ASCO 2022

DCIS assays and limitations: DCISionRT and Oncotype DX DCIS aim to stratify DCIS recurrence risk and radiotherapy benefit; current evidence is limited and clinical‑outcome data from score‑guided decisions are lacking

ESMO 2024 note: When uncertainty about adjuvant chemo remains after clinicopathologic assessment, GEP and endocrine response assessment can be used to assist individualized decisions

ESMO 2024

Specimen quality requirement: Specimens must meet quality-control standards; archived tissue may fail QC and limit reporting

Document specimen type and QC

CUP molecular testing limitations: Molecular tests to direct site‑specific therapy in cancers of unknown primary currently lack consistent randomized evidence of improved outcomes; CGP/NGS may identify actionable alterations for targeted therapy or trials but should be used after standard clinicopathologic workup

NCCN/ESMO guidance

Society recommendations: Guidelines support targeted somatic profiling in pancreatic cancer when patients are candidates for therapy and recommend germline testing for confirmed pancreatic cancer; do not endorse broad screening with MCED tests

NCCN and ASCO positions

The policy lists other breast GEP assays and related profiling tests (for example, 41‑gene signatures, 76‑gene Rotterdam, DCISionRT and others not previously described) as lacking sufficient evidence; these assays are presented as examples of tests for which clinical utility has not been established.

For DCIS prognostic assays, Hayes and systematic reviews concluded evidence quality is low and clinical utility is uncertain. Specifically, Oncotype DX DCIS has a low‑quality evidence base and the clinical outcomes from DCIS score–based decision‑making are unknown; NICE also placed restrictions on use of some assays (e.g., MammaPrint) for guiding adjuvant chemotherapy in selected node‑positive groups.

Hayes molecular test assessments for prostate genomic assays concluded that evidence is frequently low quality and insufficient to draw definitive conclusions about clinical utility in certain prostate cancer populations, particularly lower‑risk localized disease and some higher‑risk groups; these findings support conservative coverage positions pending stronger outcome data.

The policy identifies several tests judged by evidence reviews to have insufficient clinical validity or utility (examples include Select mdx, Confirm mdx, ExoDx); independent assessments found the evidence base inadequate to support routine clinical use or coverage for these assays.

The policy states that molecular biomarkers without sufficient supporting data or that are not commercially available should not be offered routinely. Reviews (Hayes, professional society statements) concluded insufficient evidence for some newer biomarkers and recommended against offering them outside validated use cases.

Independent evidence assessments (e.g., Hayes) found that DecisionDx‑Melanoma and similar prognostic melanoma assays currently have a low‑quality evidence base and no demonstrated clinical utility for improving outcomes; as a result, these assays are not supported for routine use to guide management decisions.

Prognostic gene‑expression tests for cutaneous and uveal melanoma (including DecisionDx‑UM) lack sufficient evidence that their results change management in ways that improve outcomes. Hayes and guideline summaries conclude clinical utility is not established, so routine coverage for prognostic use in melanoma is not supported outside well‑defined research or validated protocols.

Molecular tests intended to direct site‑specific therapies for cancers of unknown primary (CUP) are not supported by sufficient peer‑reviewed evidence demonstrating improved outcomes. Specifically, the CancerTYPE ID GEP test has been assessed as having insufficient evidence to conclude effectiveness for tissue‑of‑origin identification and for leading to beneficial, site‑specific treatment decisions.

The Hayes assessment for CancerTYPE ID concluded there is insufficient evidence to support its effectiveness for identifying a site of origin in indeterminate cancers; peer‑reviewed validation and outcome studies are lacking.

Professional guideline panels (ACG) and meta‑analyses report that blood‑based SEPT9 testing (mSEPT9/Epi proColon/ColoHealth) has limited sensitivity, particularly for early‑stage disease, and is not considered an optimal replacement for standard colorectal cancer screening; guideline recommendations advise against routine use for primary screening.

The policy states that PancraGEN and other pancreatic molecular tests lack sufficient peer‑reviewed evidence to support use for routine risk assessment or to guide surgical decisions. Likewise, multi‑cancer early detection (MCED/MCD) cfDNA‑based screening assays currently lack prospective population‑level evidence of clinical utility and are not supported for routine screening.

Whole exome, whole genome, and whole‑transcriptome sequencing for tumor profiling are identified as lacking sufficient evidence to inform routine clinical decision‑making; the policy does not support routine use of these broad sequencing approaches for tumor profiling at this time.

Tumor‑informed MRD assays such as Signatera are described as having insufficient peer‑reviewed evidence demonstrating that routine use for surveillance or monitoring improves outcomes. Hayes and other reviews conclude that evidence is immature and routine surveillance use is not supported outside of clinical trials.

Large panels (example)51+ genes — represented by CPT descriptors 81455/81456 for 51+ gene panels

Vendor/PLA extremes (examples)Reported PLA/Category U codes include panels up to 521 genes (0585U) and ~600+ genes (0538U), reflecting very large targeted/CGP offerings

Specialty descriptorsPLA/Category III/other U‑codes list many specific panel sizes and content (e.g., 23‑gene NSCLC panel 0022U; 112‑gene thyroid panel 0026U)

DNA methylation sites for CNS methylation profiling

CNS methylation profiling requirementCPT/PLA descriptor (81524) specifies DNA methylation analysis of at least 10,000 methylation sites for central nervous system tumor classification

Specimen type notedAnalysis performed using DNA extracted from formalin-fixed tumor tissue (FFPE) per code description

Output describedAlgorithm(s) report probability of matching a reference tumor family and class; MGMT promoter methylation status included if performed

Common relevant codes include (but are not limited to): 81462–81464 (cfDNA panels), 81479 (unlisted molecular pathology), 81504, 81518–81521, 81552, 81599, G0327.
Map the performed assay to the specific CPT/HCPCS/PLA code; avoid using miscellaneous or non-applicable codes.

Prior Authorization

Prior Authorization Expectations for Genomic Assays

Payers may require prior authorization and pre-test documentation for genomic assays, multigene panels, tumor molecular profiling (tissue NGS/CGP), ctDNA/MRD assays, and novel or multigene prognostic classifiers. Provide clinical rationale linking the test to anticipated management changes.

Justify ordering of multigene panels, Immunoscore, post-surgical ctDNA, and other novel assays due to uncertain clinical value and potential denial risk.
For CGP/NGS in Cancer of Unknown Primary (CUP): document prior histologic workup, why site-specific testing is needed, and why tissue testing is not feasible if ordering plasma-based NGS.

Prior Authorization

Prior Authorization Recommended for Genomic Classifiers

For genomic classifiers and prognostic GEPs, prior authorization or additional documentation is often recommended to confirm guideline‑concordant indication, patient risk group, and that results will impact management decisions.

Document clinical indication, risk group (e.g., nodal status, stage), and how the assay result will change treatment or surveillance.
Phrases such as “test will inform decision regarding adjuvant chemotherapy” or “test will inform extended endocrine therapy (EET) decision” help demonstrate medical necessity.

Note

Selective Ordering Recommended

Order molecular biomarkers selectively—only when results, together with routine clinical factors, are reasonably likely to influence management. Routine or reflex ordering without a management intent increases denial risk.

Do not routinely order tissue-based genomic biomarkers for all patients; restrict to scenarios where added risk stratification may change therapy.
If multiple predictive GEPs are considered for the same breast tumor, only one is medically necessary (BCI exception noted).

Documentation Required

Selection & Counseling for Thyroid Molecular Testing

When considering molecular testing for indeterminate thyroid nodules, select the test that addresses the specific clinical question and counsel patients on benefits, limitations, and uncertainties. Consider repeat FNA before molecular testing when appropriate.

If thyroidectomy is already indicated for clinical reasons, molecular testing is unnecessary.
For Bethesda III (AUS/FLUS) nodules, consider repeat FNA or molecular testing depending on clinical, sonographic features and patient preference.
Document pre-test counseling about test limitations and potential impact on surgical decision-making.

Prior Authorization

Prior Authorization Considerations for CGP/NGS in CUP

For CGP/NGS in CUP cases, prior authorization may require documentation of exhaustive histologic/immunohistochemical workup and rationale for genomic profiling. Tissue-based testing is preferred; plasma cfDNA may be acceptable only if tissue is unavailable or inadequate.

Provide prior diagnostic workup notes (histology, IHC) and explain how CGP results would change therapy or trial eligibility.
If ordering plasma NGS for CUP, document why tissue testing is not feasible and the planned actionable use of identified alterations.

Prior Authorization

Prior Authorization for Tumor Molecular Profiling

Tumor molecular profiling (tissue NGS/CGP) commonly requires pre-authorization and thorough documentation of intended use—demonstrate that testing will guide targeted therapy in advanced/metastatic disease or affect clinical management per guidelines.

For pancreatic adenocarcinoma, NCCN supports somatic profiling in metastatic/locally advanced disease when patient is a candidate for therapy—document candidacy.
Provide specimen details (FFPE tissue, biopsy site, adequacy) and testing platform/genes interrogated.

Prior Authorization

Prior Authorization Advised for Signatera and Similar MRD Assays

For tumor-informed and tumor‑naive MRD assays (e.g., Signatera), payers may request prior authorization and will expect documentation of the clinical context and intended management change; routine use for surveillance is not currently supported by sufficient evidence.

Hayes and NICE found insufficient evidence that Signatera improves outcomes or guides therapy; prior authorization should include rationale and plan for action if positive.
Document the surveillance plan and how positive or negative MRD results will change imaging, systemic therapy, or enrollment in a clinical trial.

Note

No Explicit Universal Prior Authorization; Verify Member Benefit

No single explicit universal prior authorization policy text appears in the source—coverage and prior authorization decisions depend on the member-specific benefit plan and submitted documentation.

Verify member benefit and follow the payer's authorization workflow; absence of explicit policy language here does not imply no authorization is required.
Failure to provide requested medical records or justify medical necessity can lead to claim denial.

Denial Risk

Denial Risk from Limited or Low-Quality Evidence

Insufficient or low-quality evidence, limited clinical utility, or use outside guideline‑recommended indications increases the risk of claim denial. Tests with uncertain impact on long‑term outcomes (e.g., many prognostic GEPs, some MRD assays) are particularly vulnerable.

Document guideline concordance when available (NCCN, ASCO, AUA/ASTRO, ATA, NICE).
Tests lacking prospective outcome data or clear actionability should include strong clinical justification to reduce denial risk.

Denial Risk

Evidence-Based Denial Risk & Insufficient Clinical Utility

Some tests (prognostic GEPs for melanoma, certain prostate and other prognostic panels) are discouraged for routine use due to insufficient clinical utility evidence. Prior authorization reviewers may deny testing when evidence does not support clinical benefit.

For cutaneous melanoma, prognostic GEP testing is discouraged outside clinical studies; document multidisciplinary rationale if considered.
For prostate genomic tests, evidence quality varies; document how the result will alter management (e.g., active surveillance vs definitive therapy).

Note

Adherence to Guideline Sequencing & Indications

Adhere to guideline sequencing and indications (e.g., ASCO, NCCN, NICE). Do not order tests outside their intended use or recommended populations—for example, follow ASCO/NICE recommendations for breast GEP use and NCCN sequencing for prostate, colon, pancreatic, and thyroid testing.

Oncotype DX, MammaPrint, BCI, EndoPredict have specific recommended indications—document menopausal status, nodal status, and how results will inform adjuvant chemotherapy or EET.
NCCN recommends universal MMR/MSI testing for colorectal cancer and somatic profiling for metastatic pancreatic cancer—document adherence.

Denial Risk

Risk of Denial for SEPT9-Based or Non‑Standard Screening

Certain screening or blood-based tests (e.g., SEPT9 methylation tests for colorectal cancer) have variable performance and are not recommended as replacements for guideline screening; document rationale if using such tests.

NCCN notes SEPT9 performance may be reduced in early-stage disease and affected by comorbidities (age, diabetes, arteriosclerosis, arthritis).
Be aware of repeat-interval uncertainty and potential payer denial for substituting established screening modalities.

Documentation Required

Specimen, Test Description & Laboratory Requirements

Document specimen type, quality, and the test methodology. Labs should participate in external quality assurance (e.g., CLIA/CAP) and testing should be performed within the intended use of the assay.

Specify specimen (FFPE tissue block, fresh frozen tissue, plasma cfDNA), tumor fraction, and date of collection.
Provide lab accreditation status and note if archived/older tissue failed QC or required repeat sampling.

Documentation Required

EndoPredict & Breast GEP Documentation

For EndoPredict and other breast GEPs, include required clinical variables (ER/PR status, HER2, nodal status, stage) and document that results will be used for decisions about adjuvant chemotherapy or extended endocrine therapy.

When ordering EndoPredict, document hormone receptor status, HER2 status, nodal status, and whether the test result will influence chemo vs endocrine therapy decisions.
Note any prior GEP testing to avoid medically unnecessary repeat testing (BCI exception applies).

Documentation Required

Shared Decision‑Making Documentation

Shared decision-making should be documented: clinicians must explain the meaning of tumor profiling results, associated risks and benefits, and how results inform management. Document the patient‑clinician discussion.

Record that the patient was counseled on possible outcomes and how each result could alter adjuvant therapy, surveillance, or surgery.
Include patient preferences and consent for testing in the medical record.

Documentation Required

Required Clinical Documentation & Management Intent

Required clinical documentation to support coverage decisions includes disease stage, nodal status, PSA/Gleason/clinical T stage (when applicable), biopsy core data, prior treatments, and explicit management intent tied to test results.

For prostate assays: provide PSA, Gleason score, clinical T stage, percent positive cores, and NCCN risk group.
For breast assays: document menopausal status, node status, hormone receptor and HER2 status, and planned therapy pathway.

Documentation Required

Document Decision Rationale & Management Linkage

Document the decision rationale—how test results are expected to change management (e.g., enable omission of chemotherapy, guide EET, inform adjuvant radiation, select targeted therapy or clinical trial enrollment).

For cSCC 40‑GEP use to guide adjuvant radiation therapy, include multidisciplinary justification (dermatology, radiation oncology, surgery).
For CUP or advanced disease, link CGP/NGS results to potential actionable targets or trial eligibility.

Documentation Required

Clinical Candidacy Documentation

Clinical candidacy for testing must be documented—show that the patient meets guideline or evidence-based criteria for testing and that test results will be actionable.

Examples: metastatic pancreatic cancer candidate for systemic therapy (NCCN), early-stage ER+/HER2- breast cancer meeting assay indications (ASCO/NICE).
If candidacy is borderline, include multidisciplinary notes and rationale.

Documentation Required

Require Clinical Correlation with ctDNA / Plasma HPV DNA Results

For plasma ctDNA/TTMV HPV DNA tests, require clinical, radiographic, or pathologic correlates when reporting positive results. Persistent ctDNA or HPV DNA without correlates is of limited actionability and may not support treatment changes outside trials.

Document concurrent imaging or pathologic confirmation plans when ordering MRD or HPV ctDNA assays.
Note that negative ctDNA does not rule out recurrence; imaging-first surveillance per guidelines is preferred.

Denial Risk

Multiple GEP Restrictions

Multiple GEPs for the same tumor are generally not medically necessary. Do not order more than one predictive GEP for the same breast tumor (exception: BCI may be used once for EET decisions even if another GEP was previously used).

If a second GEP is requested, document the clinical justification and how the second test provides non‑redundant, actionable information.

Step Therapy

Use of GEP Results, Step Therapy & Sequencing Notes

Where evidence or guidelines allow, use GEP results before escalating therapy and document that the test informed escalation/de‑escalation decisions. No mandated step therapy or test sequencing is specified, but follow guideline-recommended sequencing when present.

No payer-mandated step therapy is specified in source — however, document adherence to guideline sequencing (e.g., Oncotype DX use in premenopausal ER+ node-negative disease).
Consider imaging and routine markers first for surveillance; ctDNA may be adjunctive but not a replacement for guideline imaging.

Documentation Required

Consider Repeat FNA Before Molecular Testing

Before ordering molecular tests for indeterminate thyroid nodules, consider and document repeat FNA when appropriate; if proceeding with molecular testing, document counseling and expected management consequences.

For Bethesda III nodules, list clinical and sonographic features considered and whether repeat FNA was performed or declined.
If molecular testing supports observation vs surgery, document the planned management contingent on test results.

Documentation Required

Multidisciplinary Justification for ART and Complex Decisions

For decisions about adjuvant radiation therapy or complex treatment choices (e.g., using 40‑GEP in cSCC), include multidisciplinary justification in the record to support medical necessity and help authorization reviewers.

Document input from relevant specialists (e.g., dermatology, radiation oncology, surgical oncology) and how the genomic classifier result influenced therapy selection.

Note

Imaging‑First Surveillance Preferred; ctDNA as Adjunct

Imaging-first surveillance per guideline recommendations (e.g., NCCN) remains preferred. ctDNA or MRD assays should be considered adjunctive and ordered only when results will change subsequent management and are accompanied by a plan for confirmatory imaging or pathology.

When using ctDNA for surveillance, document the plan for confirmatory imaging and the intended management pathway if ctDNA becomes positive.
Avoid relying solely on ctDNA to make treatment changes absent radiographic or pathologic corroboration.

Performed on tumor tissue (often FFPE); panel sizes vary from targeted gene sets to hundreds of genes depending on test

RoleUsed to identify actionable alterations for targeted therapy selection or clinical trial eligibility when clinically indicated

Liquid Biopsy — definition and uses

DefinitionLiquid biopsy: testing of bodily fluids (commonly plasma) to identify tumor cells or circulating tumor DNA/RNA for detection, treatment selection, or monitoring

ApplicationsUsed as a noninvasive alternative or complement to tissue profiling in NSCLC and other settings when tissue is limited

LimitationsAnalytic and clinical performance vary; cfDNA/ctDNA sensitivity can be affected by tumor fraction and disease burden — correlate with clinical context

ctDNA / cell-free DNA — definition and purposes

DefinitionctDNA / cell-free DNA are fragments of tumor-derived DNA circulating in plasma used to detect sequence variants, methylation patterns, tumor fraction, or minimal residual disease (MRD)

PurposesUsed for molecular characterization, detection of actionable mutations, monitoring treatment response, and investigational MRD surveillance

DocumentationClinical, radiographic, or pathologic correlates should accompany ctDNA results due to limitations in actionability when used alone

FFPE — specimen type definition

DefinitionFFPE (formalin-fixed paraffin-embedded) denotes tumor tissue fixed and embedded for histology; commonly specified specimen type for tumor DNA/RNA assays

ImplicationArchived/older FFPE specimens may fail quality-control and affect ability to report valid genomic results

UseMost tissue-based GEP and NGS assays accept FFPE; document specimen type when ordering and for authorization workflows

Multi-gene analysis — panel definition

DefinitionMulti‑gene analysis generally refers to a gene panel containing five or more genes

Threshold used in policyPolicy treats panels with ≥5 genes as multi‑gene; coding descriptors (e.g., 81445) align with 5–50 gene panel definitions

RelevanceGuides selection of appropriate CPT/PLA codes and interpretation of coverage rules (e.g., small panels vs CGP)

Gene-expression assay — definition and examples

DefinitionGene‑expression assay (GEP) measures tumor mRNA expression patterns to generate prognostic or predictive scores (examples include Oncotype DX 21‑gene, Prosigna PAM50 50‑gene)

ExamplesOncotype DX (21 genes), Prosigna/PAM50 (50 genes), MammaPrint (70 genes), EndoPredict (12 genes) — used for prognosis and sometimes to inform therapy

SpecimenTypically measured from FFPE tumor tissue; some assays accept fresh/frozen tissue per test instructions

Prosigna (PAM50)

DefinitionProsigna (PAM50) is a 50‑gene expression assay performed on FFPE tumor tissue that provides a risk category and numerical score estimating ten‑year distant recurrence risk in postmenopausal HR+ N0 or N1 breast cancer

LimitationNot intended for individuals with N2 (four or more positive nodes)

OutputProvides intrinsic subtype and risk of distant recurrence to inform prognosis and treatment planning

MammaPrint (70-gene signature)

DefinitionMammaPrint is a 70‑gene expression signature performed on tumor tissue (fresh, frozen, or FFPE) that classifies tumors as low‑risk or high‑risk for distant recurrence

UseMay be used to inform adjuvant systemic therapy decisions in selected early breast cancer populations (per guideline context)

Often pairedOften used with BluePrint molecular subtyping; clinical utility of BluePrint in isolation is limited per policy

BluePrint (80-gene)

DefinitionBluePrint is an 80‑gene molecular subtype test designed to classify breast tumors into one of three subtypes to predict behavior and therapy response, often used with MammaPrint

Evidence noteEvidence for clinical utility of BluePrint in conjunction with MammaPrint is described but considered insufficient in some contexts

SpecimenMeasured from tumor tissue (FFPE or fresh/frozen depending on test platform)

Oncotype DX Recurrence Score (RS)

DefinitionOncotype DX Recurrence Score (RS) is a 21‑gene multigene RT‑PCR assay that provides a recurrence score to stratify distant recurrence risk and inform adjuvant chemotherapy decisions in ER+/HER2- early breast cancer

Score horizonRecurrence risk typically referenced over a ten‑year horizon in breast cancer studies and guidelines

Clinical rolePreferred by NCCN for prognosis and predicting chemotherapy benefit in selected patients

EPclin Risk Score (EndoPredict)

DefinitionEPclin Risk Score (EndoPredict) integrates a 12‑gene signature with tumor size and nodal status to produce a numeric score (approximately 1.1–6.2) estimating recurrence risk

ComponentsIncludes 8 disease‑relevant genes, 3 RNA normalization genes and 1 DNA reference gene; combines molecular and clinicopathologic data

Clinical noteProspective and retrospective studies report changes in adjuvant chemotherapy recommendations based on EPclin results, but additional validation is needed

DCISionRT DS

DefinitionDCISionRT Decision Score (DS) is a 0–10 score combining a seven‑gene expression result with clinicopathologic factors to estimate ten‑year DCIS recurrence risk and potential radiotherapy benefit

Risk categoriesScores 0–3 considered low risk; 3–10 considered elevated risk

EvidenceSystematic reviews and assessments note limited and mixed evidence on clinical outcomes from DS‑guided decisions

Oncotype DX DCIS Score

DefinitionOncotype DX DCIS Score is a 12‑gene RT‑PCR assay producing a 0–100 score to represent ten‑year risk of DCIS recurrence

Intended useAims to inform recurrence risk and potentially guide radiotherapy decisions in DCIS, though evidence quality is limited

Evidence cautionHayes assessment found low‑quality evidence and unknown impact on clinical outcomes for DCIS score–guided decisions

GEP assays examples summary

Summary examplesCommon breast GEP examples include 21‑gene Oncotype DX, 70‑gene MammaPrint, 50‑gene Prosigna (PAM50), and 12‑gene EndoPredict — each provides prognostic (and in some cases predictive) information

Clinical implicationsAssay choice and interpretation depend on clinical context (menopausal status, nodal status, intended management) per guidelines

Coding noteMultiple CPT/PLA descriptors exist for these assays (e.g., 81519 Oncotype DX 21‑gene; 81521/81523 MammaPrint variants; 81524 methylation profiling) — select code matching test method and content

Liquid biopsy (ctDNA/cfDNA) — noninvasive use

DefinitionLiquid biopsy (ctDNA/cfDNA) is a noninvasive alternative to tissue profiling used for molecular characterization and monitoring, particularly in NSCLC when tissue is limited

Prognostic utilityMeta-analyses show baseline ctDNA negativity and early clearance after therapy are associated with improved PFS/OS, but heterogeneity exists and further serial testing studies are recommended

Operational noteInterpretation should consider tumor fraction, assay sensitivity, and need for clinical correlation

Decipher Prostate Biopsy GC

DefinitionDecipher Prostate Biopsy GC is a whole transcriptome‑based genomic classifier evaluating expression of 22 genes from biopsy tissue to estimate risk of adverse outcomes and assist treatment decisions

Use contextsUsed for risk stratification at diagnosis and to inform post‑prostatectomy adjuvant therapy decisions in select cases (NCCN category 2B for post‑RP use)

Evidence noteHayes found variable quality evidence; associations with outcomes exist but impact on long‑term outcomes is uncertain

GPS (Genomic Prostate Score / Oncotype DX)

DefinitionGPS (Genomic Prostate Score / Oncotype DX Prostate) is a 17‑gene expression score measured on biopsy tissue intended to predict risk of adverse pathology and inform management decisions

Clinical roleIntended to assist decisions about active surveillance versus definitive therapy in select localized prostate cancer patients

EvidenceMeta-analyses show higher GPS associated with increased risk of metastasis and recurrence, though evidence for improved outcomes from GPS‑guided care is limited

Prolaris (CCP)

DefinitionProlaris (CCP) is a cell‑cycle progression gene‑expression profile used to predict tumor aggressiveness and assist decisions between active surveillance and definitive treatment

Assay typeMeasures expression of cell‑cycle related genes to produce a score correlated with adverse pathology and outcomes

Clinical noteUsed selectively per guideline recommendations; evidence varies across risk groups

Genomic expression classifiers (examples)

DefinitionGenomic expression classifiers (GECs) are targeted gene‑expression tests (examples: Oncotype DX/GPS, Prolaris, Decipher, ProMark) used to stratify risk and inform management decisions

Common usesRisk stratification at diagnosis, informing active surveillance vs treatment, and select post‑operative adjuvant decisions

Evidence considerationsGuidelines recommend selective use; routine ordering is not advised without demonstrated impact on outcomes

NCCN prostate cancer risk groups

DefinitionNCCN prostate cancer risk groups classify localized prostate cancer into very low, low, intermediate (favorable/unfavorable), high, and very high risk based on clinical stage, Grade Group, PSA, and biopsy core involvement

Use in testingRisk group assignment is used to determine appropriateness of genomic testing (e.g., GPS, Decipher, Prolaris) and management pathways

Clinical implicationAssay results should be interpreted in the context of NCCN risk group and life expectancy when informing treatment

Bethesda III/IV — indeterminate thyroid nodules

DefinitionBethesda III/IV denotes indeterminate thyroid cytology categories for fine‑needle aspiration where molecular testing is commonly used to assess malignancy risk

Testing roleMolecular panels (ThyroSeq), Afirma GSC/XA, and miRNA classifiers (ThyraMIR) are applied in this population to help avoid unnecessary diagnostic surgery when results influence management

Performance noteMeta‑analyses report high pooled sensitivity and NPV for some tests (e.g., ThyroSeq v3, Afirma GSC) but specificity and PPV vary; interpret with clinical context

ThyraMIR — miRNA classifier

DefinitionThyraMIR is a miRNA‑based classifier used alongside mutation panels and mRNA sequencing to improve sensitivity and negative predictive value for indeterminate thyroid nodules

Use caseOften combined with ThyroSeq/ThyGeNEXT panels to increase rule‑out performance and avoid unnecessary surgery

EvidenceStudies show addition of miRNA classifiers can increase sensitivity and NPV in some cohorts, but prospective outcome data remain limited

31-GEP / i31-GEP

Definition31‑GEP / i31‑GEP is a 31‑gene expression profile (DecisionDx/DecisionDx‑Melanoma) intended to predict sentinel lymph node positivity and recurrence risk in cutaneous melanoma

Clinical applicationMay influence decisions regarding sentinel lymph node biopsy (SLNB) in T1‑T2 tumors and has been associated with reductions in SLNB rates in prospective studies

Evidence cautionStudies show reduced SLNB rates but limitations and potential bias (manufacturer involvement) warrant cautious interpretation

Pigmented Lesion Assay (PLA)

DefinitionPigmented Lesion Assay (PLA) is a noninvasive tape‑stripping test (DermTech) that measures gene expression (e.g., LINC00518, PRAME, TERT) to help rule out melanoma and guide biopsy decisions

Clinical noteMay reduce unnecessary biopsies in some studies but real‑world discordance with histopathology and analytic failures have been reported

Evidence limitationHayes and HTA reviews note low‑quality evidence and recommend further study before routine adoption

31-gene GEP

Definition31‑gene GEP (cutaneous melanoma) is a classifier that categorizes tumors into risk classes (e.g., Class 1 low risk, Class 2 high risk) to prognosticate metastasis and recurrence

Clinical outputProvides risk class associated with differences in recurrence‑free and distant metastasis‑free survival in cohort studies

LimitationsProspective validation and demonstration of improved patient outcomes from GEP‑guided management remain limited

15-gene GEP / DecisionDx-UM

Definition15‑gene GEP (DecisionDx‑UM) is a prognostic assay for uveal melanoma that stratifies patients into low, intermediate, and high metastatic risk groups to guide surveillance intensity

Clinical impactProspective registry data show class‑specific management intensity and differing metastasis‑free survival rates, but manufacturer sponsorship raises potential bias concerns

Use conditionNCCN allows prognostic stratification if biopsy is performed; decision to test should consider how results will alter follow‑up

40-GEP / DecisionDx-SCC

Definition40‑GEP / DecisionDx‑SCC is a 40‑gene expression profile for cutaneous squamous cell carcinoma that categorizes patients into Class 1, 2A, or 2B risk groups for metastasis to inform adjuvant radiation decisions

ValidationProspective multicenter and validation cohorts report differing metastasis‑free survival by class and a PPV of ~60% for highest‑risk class (2B)

GuidanceSome multidisciplinary guidelines discuss use in ART decision‑making but recommend multidisciplinary justification due to mixed evidence and potential manufacturer involvement

PLA (DermTech)

DefinitionPLA (DermTech) is the tape‑stripping Pigmented Lesion Assay intended to help rule out melanoma and inform biopsy decisions for suspicious melanocytic lesions

LimitationsAnalytic failures and discrepancies with histopathology reported; clinical utility remains uncertain per HTA assessments

Operational noteUse as adjunct to clinical and dermatopathologic evaluation rather than standalone diagnostic tool

CGP/NGS — definition

DefinitionCGP/NGS refers to broad next‑generation sequencing approaches that assess multiple classes of genomic alterations (SNVs, indels, CNAs, fusions, MSI, TMB) often across hundreds of genes

PurposeUsed to identify actionable alterations for targeted therapy or clinical trial enrollment in advanced/metastatic disease or CUP workups

Specimen and code relevancePerformed on FFPE tissue or plasma cfDNA; large panels are captured in PLA/U‑code descriptors (e.g., 0244U, 0250U, 0538U)

GEP — gene expression profiling definition

DefinitionGEP (gene expression profiling) tests analyze RNA expression signatures to predict tumor behavior or tissue of origin and can be used diagnostically, prognostically, or predictively

MethodsMethods include RT‑PCR, microarray, and targeted RNA‑seq; selection depends on assay and intended output

ApplicationsUsed in breast cancer prognostication, melanoma risk stratification, and CUP tissue‑of‑origin classification (limited outcome evidence)

These excerpts did not contain top‑level eligibility requirement nodes in the provided segments.

Cell‑free DNA panel examplescfDNA panels described with 55–74 genes (0242U), 83+ genes (0326U), or 311+ genes (0239U) in PLA descriptors — gene counts vary by vendor

Implication for authorizationDocument exact panel gene count and clinical rationale when submitting authorization for panels at or above 51 genes

DNA methylation sites threshold — coding detail

Technical thresholdCNS methylation profiling CPT/PLA descriptor (81524) specifies analysis of at least 10,000 methylation sites — use this code when test meets that analytic scope

Specimen requirementDescriptor specifies DNA extracted from formalin‑fixed tumor tissue (FFPE)

Report contentAlgorithms report probability of matching a reference tumor family/class and may include MGMT promoter methylation status if performed