Bio Clinical Databases Tumor Mutational Burden

Version Compatibility

Reference examples tested with: cyvcf2 0.30+, VEP 111+ (or snpEff 5.2+), pandas 2.2+, numpy 1.26+, LOHHLA 1.0+ (Marty 2017), DASH 1.0+ (Montesion 2021). v4.1 (May 2024) gnomAD is current for germline subtraction. Friends of Cancer Research TMB harmonization framework (Vega 2021 Ann Oncol) and ESMO 2024 (Mosele Ann Oncol) define the operational thresholds.

Before using code patterns, verify installed versions match. If versions differ:

• Python: pip show <package> then help(module.function) to check signatures
• CLI: <tool> --version

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying. TMB calculation requires VCF with VEP / snpEff / Funcotator consequence annotations; the panel size used as denominator MUST match the assay's actual scored region, NOT the panel's total content.

Tumor Mutational Burden; Calculation, Harmonization, ICI Eligibility

'Calculate TMB from this somatic VCF and apply ICI eligibility cutoff' -> Count nonsynonymous coding variants passing VAF/depth/germline filters; divide by assay scored region in Mb; apply assay-calibrated TMB-H cutoff; integrate with MSI / HLA-LOH / neoantigen quality.

• Python: cyvcf2.VCF() + VEP/snpEff consequence parsing + panel-size normalization
• CLI: bcftools view filtering + custom counting
• HLA-LOH: LOHHLA (Marty 2017 Cell) or DASH (Montesion 2021 Cancer Discov)
• Neoantigen quality: pVAC-tools, NetMHCpan-4.1, Luksza 2017 fitness model

Regulatory and Trial Landscape

Friends of Cancer Research Harmonization: Cross-Panel Calibration

Merino 2020 J Immunother Cancer: in silico panel sampling from TCGA WES truth showed panel-specific TMB can differ 3-fold for identical samples. Vega 2021 Ann Oncol derived per-panel calibration equations to translate panel TMB to WES-aligned TMB-Mb.

Per-panel calibration to FoundationOne 10/Mb sensitivity:

TMB =/= TMB across vendors. Manuscripts that compare TMB across panels without per-assay calibration are unreviewable. Use the Vega 2021 calibration equations or WES re-projection.

Variant-Counting Subtleties

These choices alter TMB by 5-20%:

Hypermutator Tiering

MSI-H typically 30-50 mut/Mb; pure POLE-exo P286R 100-300 mut/Mb; POLE-exo + MMR-D exceeds 500. MSI-H and TMB-H overlap substantially in CRC and endometrial (~80% of MSI-H are TMB-H) but only ~16% of TMB-H solid tumors are MSI-H (Salem ME et al 2018 Mol Cancer Res 16:805-812).

The Tumor-Type-Specific Cutoff Debate

McGrail 2021 Ann Oncol is the most damning paper for the universal 10/Mb cutoff. TMB-H predicts ICI response in melanoma, NSCLC, bladder; but FAILS in breast, prostate, glioma. ORR in TMB-H melanoma/NSCLC/bladder was 39.8%; TMB-H breast/prostate/glioma was 15.3%. Mechanistic explanation: TMB only predicts when baseline CD8 T-cell infiltrate is present.

Sha 2020 Cell Rep Med: TMB-H predicts ICI benefit in MSS subset but adds nothing on top of MSI-H (because MSI-H is uniformly hypermutator and uniformly responsive).

Samstein 2019 Nat Genet (MSK-IMPACT 1,662 ICI-treated): cancer-specific TMB cutoffs (top 20% within each tumor type) outperform universal 10/Mb.

ESMO 2024 retained TMB-H >= 10/Mb pan-tumor but explicitly noted exceptions: NOT endorsed for breast, prostate, glioma based on negative real-world data.

Blood TMB (bTMB): The Negative-Trial Story

Gandara 2018 Nat Med: bTMB on Foundation Medicine FoundationACT panel; POPLAR + OAK retrospective. bTMB >= 16 mut/Mb showed PFS benefit with atezolizumab in NSCLC.

B-F1RST (Kim 2022): bTMB >= 16 prospectively predictive for atezolizumab first-line NSCLC.

BFAST Cohort C (Dziadziuszko 2022): FAILED primary endpoint; atezolizumab vs chemo in bTMB-H NSCLC did not improve investigator-assessed PFS. Dominant confounder: low ctDNA shed fraction produces false-negative bTMB.

Operational state: bTMB is research-grade in tissue-naive settings; tissue TMB remains the regulatory standard.

Neoantigen Quality: Beyond Raw TMB

Luksza 2017 Nature: neoantigen fitness model. Combines "non-selfness" (TCR recognition probability via IEDB similarity) + "selfness" (MHC binding affinity differential vs WT peptide). Pancreatic-cancer validation (Balachandran 2017 Nature): long-term survivors had higher-quality neoantigens. Luksza 2022 Nature: immunoediting over 10 years.

McGranahan 2016 Science: clonal neoantigen burden (mutations present in all tumor cells) predicts ICI response better than total. Subclonal-rich tumors evade despite high TMB.

HLA-LOH (Marty 2017 Cell, LOHHLA; Montesion 2021 Cancer Discov, DASH): HLA-LOH occurs in ~40% of NSCLC and abolishes neoantigen presentation for the lost allele. ~17% pan-cancer; >30% in HNSCC / NSCLC / cervical. Co-occurs with high subclonal burden + APOBEC + immune escape.

Decision Tree by Scenario

Standard Workflow

Goal: Compute TMB from a VEP-annotated somatic VCF with full filtering.

Approach: Parse cyvcf2; apply VAF + depth + germline (gnomAD) filters; count nonsynonymous coding consequences; divide by scored Mb.

from cyvcf2 import VCF
import re

NONSYNONYMOUS_CONSEQUENCES = {
    'missense_variant', 'stop_gained', 'stop_lost', 'start_lost', 'start_retained',
    'frameshift_variant', 'inframe_insertion', 'inframe_deletion',
    'splice_donor_variant', 'splice_acceptor_variant',
    'protein_altering_variant', 'initiator_codon_variant'
}

# Vega 2021-calibrated scored regions (Mb)
PANEL_SCORED_REGION = {
    'FoundationOne_CDx': 0.8,           # Scored region; NOT 1.1 panel total
    'MSK_IMPACT_v3': 0.98,
    'MSK_IMPACT_v4': 1.22,
    'TSO500': 1.3,                       # Scored from 1.94 total
    'Oncomine_TML': 1.2,
    'Caris_MI': 1.2,
    'Tempus_xT_v3': 0.6,                 # Borderline reliability
    'WES': 30.0,
    'WGS': 3000.0
}

# Vega 2021 calibration: equivalent thresholds for FDA 10/Mb FoundationOne sensitivity
ASSAY_TMB_H_CUTOFF = {
    'FoundationOne_CDx': 10.0,
    'TSO500': 7.8,
    'Oncomine_TML': 8.4,
    'MSK_IMPACT_v3': 10.0,  # Approximate; full Vega 2021 calibration recommended
    'MSK_IMPACT_v4': 10.0,
    'WES': 10.0
}


def parse_consequences_from_vep(csq_field, csq_header):
    '''Parse VEP CSQ INFO field; returns list of per-transcript consequence types.'''
    if not csq_field:
        return []
    cons_idx = csq_header.index('Consequence')
    out = []
    for transcript in csq_field.split(','):
        fields = transcript.split('|')
        if len(fields) > cons_idx:
            out.append(fields[cons_idx])
    return out


def is_nonsynonymous(consequences, include_synonymous=False):
    '''Check if variant has nonsynonymous coding consequence.

    FoundationOne CDx convention INCLUDES synonymous (set include_synonymous=True).
    MSK-IMPACT and most academic pipelines exclude.
    '''
    target = set(NONSYNONYMOUS_CONSEQUENCES)
    if include_synonymous:
        target.add('synonymous_variant')
    for cons_str in consequences:
        for cons in cons_str.split('&'):
            if cons in target:
                return True
    return False


def calculate_tmb(vcf_path, scored_region_mb, csq_header,
                   min_vaf=0.05, min_depth=100, max_gnomad_af=0.005,
                   include_synonymous=False, exclude_hotspots=True,
                   hotspot_bed=None):
    '''Calculate TMB with filtering per Vega 2021 harmonization.

    Args:
        scored_region_mb: panel's SCORED region (NOT total panel)
        min_vaf: 0.05 (FoundationOne) to 0.10 (tumor-only no UMI)
        max_gnomad_af: 0.005 (0.5%) typical for tumor-only germline filter
        include_synonymous: True for FoundationOne CDx-compatible; False for MSK-IMPACT
        exclude_hotspots: COSMIC drivers excluded (not random mutations)
    '''
    vcf = VCF(vcf_path)
    cons_idx = csq_header.index('Consequence') if 'Consequence' in csq_header else 1
    nonsyn_count = 0
    total_pass = 0

    for v in vcf:
        if v.FILTER is not None:  # FILTER == None means PASS in cyvcf2
            continue
        depth = v.INFO.get('DP', 0)
        if depth < min_depth:
            continue
        vaf = _get_vaf(v)
        if vaf is None or vaf < min_vaf:
            continue
        gnomad_af = v.INFO.get('gnomAD_AF', 0) or v.INFO.get('AF_popmax', 0) or 0
        if gnomad_af > max_gnomad_af:
            continue
        total_pass += 1

        csq = v.INFO.get('CSQ', '')
        consequences = parse_consequences_from_vep(csq, csq_header)
        if is_nonsynonymous(consequences, include_synonymous=include_synonymous):
            nonsyn_count += 1

    tmb = nonsyn_count / scored_region_mb
    return {
        'tmb': round(tmb, 2),
        'nonsynonymous_count': nonsyn_count,
        'total_passing_filters': total_pass,
        'scored_region_mb': scored_region_mb
    }


def _get_vaf(variant):
    '''Extract VAF from genotype FORMAT fields (Mutect2 AD or AF).'''
    try:
        ad = variant.format('AD')
        if ad is not None and len(ad) > 0:
            ad0 = ad[0]
            total = sum(ad0)
            return ad0[1] / total if total > 0 else None
    except Exception:
        pass
    try:
        af = variant.format('AF')
        if af is not None and len(af) > 0:
            return float(af[0])
    except Exception:
        pass
    return None


def classify_tmb(tmb_value, assay='FoundationOne_CDx'):
    '''Apply ESMO 2024 / FDA pembrolizumab cutoff with Vega 2021 calibration per assay.'''
    cutoff = ASSAY_TMB_H_CUTOFF.get(assay, 10.0)
    if tmb_value >= 500:
        category = 'Ultra-hypermutator (>=500/Mb; POLE+MMR likely)'
    elif tmb_value >= 100:
        category = 'Hypermutator (>=100/Mb; MMR-D or POLE)'
    elif tmb_value >= cutoff:
        category = f'TMB-H (>= {cutoff}/Mb {assay}-calibrated; pan-tumor ICI eligible per FDA 2020)'
    else:
        category = 'TMB-low'
    return category

TMB-MSI Concordance and Reconciliation

Goal: When MSI-H is present, TMB-H adds no information (Sha 2020).

def tmb_msi_reconcile(tmb_value, msi_status, tumor_type=None):
    '''Reconcile TMB + MSI for ICI decision.'''
    tmb_high = tmb_value >= 10
    msi_high = msi_status == 'MSI-H'

    if msi_high:
        return ('ICI eligible by MSI-H (FDA 2017 pembrolizumab); TMB-H adds no information '
                '(Sha 2020 Cell Rep Med).')
    if tmb_high and tumor_type in ('breast', 'prostate', 'glioma'):
        return ('TMB-H present but NOT endorsed for this tumor type (ESMO 2024; McGrail 2021). '
                'Tumor-specific cutoffs recommended.')
    if tmb_high:
        return ('TMB-H pan-tumor; ICI eligible (FDA pembrolizumab 2020). '
                'Confirm baseline CD8 infiltrate; check HLA-LOH (Marty 2017 LOHHLA).')
    return 'TMB-low; MSS. Standard-of-care chemo.'

Per-Operation Failure Modes

1. Using panel total size as denominator (NOT scored region)

• Trigger: Compute TMB = nonsynonymous count / 1.1 Mb for FoundationOne.
• Mechanism: FoundationOne CDx total panel is 1.1 Mb; SCORED region (counted for TMB denominator) is 0.8 Mb.
• Symptom: TMB underestimated by ~37%.
• Fix: Use 0.8 Mb for FoundationOne CDx scored region per Vega 2021.

2. Cross-panel comparison without calibration

• Trigger: TSO500 reports TMB = 9.5; compared to FoundationOne 10/Mb cutoff.
• Mechanism: Vega 2021 showed equivalent threshold is 7.8/Mb on TSO500 (not 10/Mb).
• Symptom: TSO500 TMB-H called positive at incorrect threshold.
• Fix: Apply assay-specific Vega 2021 calibration; TSO500 cutoff = 7.8/Mb.

3. FoundationOne synonymous mis-handling

• Trigger: Compare academic pipeline (no synonymous) to FoundationOne reference (synonymous included).
• Mechanism: FoundationOne CDx counts synonymous; MSK-IMPACT and most academic pipelines exclude.
• Symptom: Academic pipeline TMB systematically lower than FoundationOne by ~10-20%.
• Fix: Match the counting convention to the comparison reference; document explicitly.

4. Tumor-only TMB inflated

• Trigger: Tumor-only WES with naive germline filter (gnomAD AF > 1%).
• Mechanism: Population-specific common variants leak through if gnomAD AF threshold not stratified by ancestry.
• Symptom: AFR/EAS patient TMB inflated 1.5-3x; misclassified as TMB-H.
• Fix: Stratify gnomAD AF by patient ancestry; use grpmax FAF95; threshold <= 0.5%.

5. bTMB applied without ctDNA shed check

• Trigger: Report bTMB low in a metastatic patient.
• Mechanism: Low ctDNA shed fraction produces false-negative bTMB (BFAST Cohort C failure mechanism).
• Symptom: Patient with high tissue TMB labeled bTMB-low; ICI not offered.
• Fix: Check tumor fraction (e.g., ichorCNA, MAF of known driver) before trusting bTMB-low; consider tissue TMB.

6. TMB-H applied to breast / prostate / glioma

• Trigger: ICI prescribed for TMB-H breast cancer based on pan-tumor approval.
• Mechanism: McGrail 2021 demonstrated TMB fails to enrich for ICI response in breast, prostate, glioma; ESMO 2024 explicitly excludes these.
• Symptom: ICI offered with low expectation of benefit; patient bears unnecessary toxicity.
• Fix: Apply tumor-type-specific cutoffs (Samstein 2019); document ESMO 2024 caveat in report.

7. Hotspots inflating TMB

• Trigger: Include BRAF V600E and KRAS G12C in TMB count.
• Mechanism: Driver hotspots are non-random; including biases TMB upward in driver-mutated samples.
• Symptom: TMB inflated in samples with strong drivers.
• Fix: Exclude COSMIC-confirmed hotspots (provide hotspot BED).

8. MSI-H -> add TMB-H -> additive ICI confidence

• Trigger: Report TMB-H as additional support for ICI in MSI-H patient.
• Mechanism: MSI-H is uniformly hypermutator + uniformly ICI-responsive; adding TMB-H is statistical tautology (Sha 2020).
• Symptom: Reviewer flag.
• Fix: Report MSI-H + TMB-H concurrence but explicitly note TMB-H is NOT additive given MSI-H.

9. Ignoring HLA-LOH

• Trigger: TMB-H + neoantigen prediction without LOH check.
• Mechanism: HLA-LOH abolishes neoantigen presentation for lost allele in ~17% pan-cancer (>30% HNSCC / NSCLC / cervical).
• Symptom: Apparent neoantigen burden inflated.
• Fix: Run LOHHLA (Marty 2017) or DASH (Montesion 2021); flag HLA-LOH-positive tumors.

Reconciliation: When Sources Disagree

Quantitative Thresholds and Conventions

Common Errors

Anticipated Reviewer Pushback

References

• Marabelle A et al. 2020. Association of TMB with efficacy of pembrolizumab in advanced solid tumours from the phase 2 KEYNOTE-158 study. Lancet Oncol 21:1353.
• Merino DM et al. 2020. Establishing guidelines to harmonize tumor mutational burden (TMB). J Immunother Cancer 8:e000147. (FoC Phase I)
• Vega DM et al. 2021. Aligning TMB analytical validation efforts: TMB harmonization Phase II project. Ann Oncol 32:1626.
• Mosele MF et al. 2024. Recommendations for the use of NGS for patients with metastatic cancers. Ann Oncol 35:588. (ESMO 2024)
• McGrail DJ et al. 2021. High TMB predicts response to immunotherapy independent of clinical or pathologic features. Ann Oncol 32:661.
• Sha D et al. 2020. TMB as a predictive biomarker in solid tumors. Cell Rep Med 1:100043.
• Samstein RM et al. 2019. TMB and survival after immunotherapy across cancer types. Nat Genet 51:202.
• Chalmers ZR et al. 2017. Analysis of 100,000 human cancer genomes reveals the landscape of TMB. Genome Med 9:34.
• Yarchoan M et al. 2017. Tumor mutational burden and response rate to PD-1 inhibition. NEJM 377:2500.
• Gandara DR et al. 2018. Blood-based TMB as a predictor of response to atezolizumab in NSCLC. Nat Med 24:1441.
• Dziadziuszko R et al. 2022. Atezolizumab vs platinum-based chemotherapy in blood-based TMB-positive NSCLC: BFAST Cohort C. Nat Med 28:2541.
• Salem ME et al. 2018. Landscape of TMB in different cancers: implications for ICI response. Cancer Discov 8:1136.
• Luksza M et al. 2017. A neoantigen fitness model predicts tumour response to checkpoint blockade immunotherapy. Nature 551:517.
• Luksza M et al. 2022. Neoantigen quality predicts immunoediting in survivors of pancreatic cancer. Nature 606:389.
• McGranahan N et al. 2016. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 351:1463.
• Marty R et al. 2017. MHC-I genotype restricts the oncogenic mutational landscape. Cell 171:1272. (LOHHLA)
• Montesion M et al. 2021. Somatic HLA class I loss is a widespread mechanism of immune evasion which refines the use of TMB as a biomarker. Cancer Discov 11:282. (DASH)
• Friends of Cancer Research TMB harmonization resources: https://friendsofcancerresearch.org/tmb/

Related Skills

• clinical-databases/somatic-signatures - Mutational signatures including HRD (PARP) and MMR-D (ICI)
• clinical-databases/msi-detection - MSI-H is the related ICI biomarker
• clinical-databases/hla-typing - HLA typing for neoantigen prediction and LOH
• variant-calling/variant-calling - Mutect2 / Strelka2 somatic upstream
• variant-calling/clinical-interpretation - ACMG / AMP cancer framework