Q (haiku): How does the mutation landscape differ between primary and metastatic samples from the same patients in the MSK-CHORD Study?

▶ read_guide
{
 "uri": "cbioportal://mutation-frequency-guide"
}

▶ read_guide
{
 "uri": "cbioportal://sample-filtering-guide"
}

◀ result
# Mutation Frequency Analysis Guide

## IMPORTANT: Reporting Mutation Frequencies
- **ALWAYS report frequencies as percentages**, not raw counts: `frequency = (altered_samples / total_profiled_samples) × 100`
- For quick frequency lookups, **prefer the TCGA Pan-Cancer Atlas study first**, then offer to expand to other studies
- When reporting across multiple studies, show **ranges** (e.g., "TP53 is mutated in 30–60% of samples") rather than a single average
- **NEVER** sum mutation events across studies to compute an aggregate frequency — this can exceed 100% due to double-counting
- Warn users that samples may overlap across cohorts (e.g., MSK studies may share patients)
- **Choose and state the counting unit**: use patient-level frequencies for prevalence/rate questions unless the user explicitly asks for samples; use sample-level frequencies when the user asks about samples.
- **For "across cancer types" questions**, jump to the [Cross-Cancer-Type Mutation Frequency](#cross-cancer-type-mutation-frequency) section below — there is one correct recipe and several common wrong ones.

## Counting Unit: Samples vs Patients

Before answering any mutation count or frequency question, decide whether the unit is samples or patients and state that choice in the answer.

| User wording | Counting unit |
|--------------|---------------|
| "prevalence", "rate", "fraction of patients", "patients with", "how common is" | Patient-level: `COUNT(DISTINCT patient_unique_id)` |
| "samples", "specimens", "biopsies", sample-level cohort composition | Sample-level: `COUNT(DISTINCT sample_unique_id)` |
| Ambiguous | Ask, or default to patient-level for prevalence/rate language and say so |

### Cross-study sample-count caveat

When an answer touches more than one study and reports a sample count, prepend a one-line caveat:

> Sample IDs are unique within cBioPortal study prefixes, not guaranteed biological-sample identifiers across studies; overlapping cohorts can count the same patient/sample more than once.

Prefer one of these safer approaches:

- Use a shipped `cancer_study_query_preferences` cohort such as `pan_cancer_tcga` or `all_studies_non_redundant`.
- Restrict to one named study.
- Aggregate by `patient_unique_id` when the biological question is patient prevalence.

## STOP rule: a frequency above 100% means your query is wrong

If your query returns a frequency over 100%, **do not try to debug or explain the data inconsistency to the user**. The cause is always one of these query bugs:

- Summing mutation events instead of `COUNT(DISTINCT sample_unique_id)` for the numerator
- Using a study-wide sample count as the denominator instead of the gene-specific profiled count
- Cross-study aggregation where the same biological sample appears under multiple `sample_unique_id` values (e.g., MSK-IMPACT and MSK-CHORD share patients)
- **Joining the profiled CTE through `gene_panel` / `gene_panel_list` without a WES branch.** `gene_panel_id = 'WES'` is *not* a row in … (35166 more chars)

▶ get_study_guide
{
 "study_id": "msk_chord_2024"
}

◀ result
# Sample and Study Filtering Guide

## Overview
Proper filtering is essential for meaningful cBioPortal analysis. This guide covers filtering by studies, sample types, and other criteria.

## Study-Level Filtering

### 1. Always Filter by Study
Every query should specify a study to ensure consistent results:

```sql
-- Always include study filtering
SELECT *
FROM your_table
WHERE
    cancer_study_identifier = 'your_study_id'
    -- Additional filters...
```

### 2. Find Available Studies

```sql
-- Discover available studies
SELECT
    cancer_study_identifier,
    name,
    description,
    type_of_cancer_id
FROM cancer_study
ORDER BY cancer_study_identifier;
```

### 3. Study Information

```sql
-- Get detailed study information
SELECT
    cs.cancer_study_identifier,
    cs.name as study_name,
    cs.description,
    cs.sample_count,
    COUNT(DISTINCT p.internal_id) as patient_count
FROM cancer_study cs
LEFT JOIN patient p ON cs.cancer_study_id = p.cancer_study_id
WHERE
    cs.cancer_study_identifier = 'your_study_id'
GROUP BY cs.cancer_study_identifier, cs.name, cs.description, cs.sample_count;
```

### 4. Find Studies by Available Data Types

Use this when the user asks *"which studies have mutation and copy-number data for X"*, *"studies with expression for Y"*, *"is there any imaging data"*, or any *"studies with Z data"* question. `cancer_study` has one precomputed sample count per data type — filter on them. No join and no schema exploration needed. These are the same counts as the portal's study list and its "Data type" filter.

**Canonical query — studies with ALL requested data types for a cancer type:**

```sql
SELECT
    cancer_study_identifier,
    name,
    sample_count,
    mutation_sample_count,
    cna_sample_count
FROM cancer_study
WHERE type_of_cancer_id = 'luad'          -- ← OncoTree code from search_oncotree
  AND mutation_sample_count > 0           -- ← one condition per required data type
  AND cna_sample_count > 0
ORDER BY sample_count DESC;
```

**Data type → column** (a count of 0 means the study doesn't have that data):

| Data type in the user's question | Column (portal "Data type" filter label) |
|---|---|
| samples in the study | `sample_count` |
| mutation / mutations | `mutation_sample_count` ("Mutations") |
| copy-number / CNA / amplification / deletion | `cna_sample_count` ("CNA") |
| structural variant / fusion | `structural_variant_sample_count` |
| RNA / mRNA / gene expression (any platform) | `mrna_expression_sample_count` — use this for "which studies have expression data" |
| specifically the portal's "RNA-Seq" filter | `rna_seq_sample_count` ("RNA-Seq") — counts only the `<study>_rna_seq_v2_mrna` list; 0 for many GDC and iAtlas studies that do have RNA-Seq |
| specifically the portal's "RNA (microarray)" filter | `mrna_microarray_sample_count` ("RNA (microarray)") — counts only the `<study>_mrna` list |
| microRNA | `mirna_sample_count` ("miRNA") |
| protein / RPPA | `rppa_sample_count` ("RPPA") |
| mas … (7083 more chars)

◀ result
# MSK-CHORD (MSK, Nature 2024)

**Study ID:** `msk_chord_2024`

## Overview
Targeted sequencing via MSK-IMPACT panels. Clinical annotations include some derived from natural language processing (denoted NLP).

**Exactly five cancer types** (`CANCER_TYPE`, patients): Non-Small Cell Lung Cancer 7,809, Colorectal Cancer 5,543, Breast Cancer 5,368, Prostate Cancer 3,211, Pancreatic Cancer 3,109. There is **no melanoma** or any other cancer type; say so up front if asked, instead of substituting another type.

**No therapy-response variable.** There is no RECIST, objective response, or best-response attribute or event. For treatment-outcome questions (e.g. immunotherapy response), say this first; the only proxies are `OS_MONTHS`/`OS_STATUS`, or NLP radiology progression events (`Diagnosis` events with `SUBTYPE = 'Progression'`, key `PROGRESSION` = Y/N/Indeterminate), in patients with `Treatment` events of the relevant `SUBTYPE` (e.g. `Immuno`: 3,341 patients). Hand off the comparison to cBioPortal group comparison / survival.

**Nearly one sample per patient: 24,950 patients / 25,040 samples.** Only 90 patients have more than one sample, and all 90 have samples from two different cancer types (second primaries); only 26 have both a `Primary` and a `Metastasis` sample. There is no meaningful same-patient (paired) primary-vs-metastasis cohort. For "same patient" / paired questions, say this up front, then offer the **unpaired** comparison of all `Primary` vs `Metastasis` samples (`SAMPLE_TYPE`), labelled as unpaired.
```sql
SELECT countIf(n > 1) AS multi_sample_patients,         -- 90
       countIf(has_p AND has_m) AS primary_and_met      -- 26
FROM (SELECT patient_unique_id, count() AS n,
             has(groupArray(attribute_value), 'Primary') AS has_p,
             has(groupArray(attribute_value), 'Metastasis') AS has_m
      FROM clinical_data_derived
      WHERE cancer_study_identifier = 'msk_chord_2024' AND attribute_name = 'SAMPLE_TYPE'
      GROUP BY patient_unique_id);
```

## Gene Panels
This study uses multiple MSK-IMPACT panel versions:
- **IMPACT341**: Earlier version, 341 genes
- **IMPACT410**: 410 genes
- **IMPACT468**: 468 genes  
- **IMPACT505**: Latest version, 505 genes

**Important:** Different samples may have different gene coverage. Always use gene-specific denominators when calculating mutation frequencies.

## Clinical Attributes - Semantic Guide

### Cancer Classification
| Attribute | Description | Values |
|-----------|-------------|--------|
| `CANCER_TYPE` | Broad cancer category | e.g., "Non-Small Cell Lung Cancer", "Breast Cancer" |
| `CANCER_TYPE_DETAILED` | Specific subtype | e.g., "Lung Adenocarcinoma", "Invasive Ductal Carcinoma" |
| `ONCOTREE_CODE` | OncoTree classification code | Standardized cancer type codes |

### Sample Information
| Attribute | Description | Values |
|-----------|-------------|--------|
| `SAMPLE_TYPE` | Sample origin | Primary, Metastasis, Local Recurrence, Unknown |
| `SAMPLE_CLASS` | Sample … (5439 more chars)

▶ clickhouse_run_select_query
{
 "query": "-- Primary vs Metastasis sample type distribution in MSK-CHORD
SELECT
    attribute_value as sample_type,
    COUNT(DISTINCT sample_unique_id) as sample_count,
    COUNT(DISTINCT patient_unique_id) as patient_count
FROM clinical_data_derived
WHERE cancer_study_identifier = 'msk_chord_2024'
  AND attribute_name = 'SAMPLE_TYPE'
GROUP BY attribute_value
ORDER BY sample_count DESC;",
 "max_rows": 100
}

◀ result
{"rows":[{"sample_type":"Primary","sample_count":15928,"patient_count":15872},{"sample_type":"Metastasis","sample_count":8878,"patient_count":8872},{"sample_type":"Unknown","sample_count":136,"patient_count":136},{"sample_type":"Local Recurrence","sample_count":98,"patient_count":98}]}

▶ clickhouse_run_select_query
{
 "query": "-- Compare mutation burden (TMB) between Primary and Metastatic samples
WITH sample_types AS (
    SELECT 
        sample_unique_id,
        MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type,
        MAX(CASE WHEN attribute_name = 'TMB_NONSYNONYMOUS' THEN CAST(attribute_value AS Float64) END) as tmb
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name IN ('SAMPLE_TYPE', 'TMB_NONSYNONYMOUS')
    GROUP BY sample_unique_id
    HAVING sample_type IS NOT NULL AND tmb IS NOT NULL
)
SELECT
    sample_type,
    COUNT(*) as samples,
    ROUND(AVG(tmb), 2) as avg_tmb,
    ROUND(MIN(tmb), 2) as min_tmb,
    ROUND(MAX(tmb), 2) as max_tmb,
    ROUND(QUANTILE(0.5)(tmb), 2) as median_tmb
FROM sample_types
GROUP BY sample_type
ORDER BY sample_type;",
 "max_rows": 100
}

◀ result
{"error_message":"Query execution failed: HTTPDriver for https://dl96orhu96.us-east-1.aws.clickhouse.cloud:8443 received ClickHouse error code 46\n Code: 46. DB::Exception: Function with name `QUANTILE` does not exist. In scope WITH sample_types AS (SELECT sample_unique_id, MAX(multiIf(attribute_name = 'SAMPLE_TYPE', attribute_value, NULL)) AS sample_type, MAX(multiIf(attribute_name = 'TMB_NONSYNONYMOUS', CAST(attribute_value, 'Float64'), NULL)) AS tmb FROM clinical_data_derived WHERE (cancer_study_identifier = 'msk_chord_2024') AND (attribute_name IN ('SAMPLE_TYPE', 'TMB_NONSYNONYMOUS')) GROUP BY sample_unique_id HAVING (sample_type IS NOT NULL) AND (tmb IS NOT NULL)) SELECT sample_type, COUNT(*) AS samples, ROUND(AVG(tmb), 2) AS avg_tmb, ROUND(MIN(tmb), 2) AS min_tmb, ROUND(MAX(tmb), 2) AS max_tmb, ROUND(QUANTILE(0.5)(tmb), 2) AS median_tmb FROM sample_types GROUP BY sample_type ORDER BY sample_type ASC. Maybe you meant: ['quantile','quantiles']. (UNKNOWN_FUNCTION) (version 26.4.1.2359 (official build))\n"}

▶ clickhouse_run_select_query
{
 "query": "-- Compare top 15 most mutated genes: Primary vs Metastatic samples
WITH sample_types AS (
    SELECT sample_unique_id,
           MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name = 'SAMPLE_TYPE'
    GROUP BY sample_unique_id
),
gene_mutations_by_type AS (
    SELECT
        st.sample_type,
        ged.hugo_gene_symbol,
        COUNT(DISTINCT ged.sample_unique_id) as altered_samples,
        COUNT(*) as total_mutation_events
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND ged.variant_type = 'mutation'
      AND ged.mutation_status != 'UNCALLED'
      AND ged.off_panel = 0
      AND st.sample_type IN ('Primary', 'Metastasis')
    GROUP BY st.sample_type, ged.hugo_gene_symbol
),
profiled_samples AS (
    SELECT 
        st.sample_type,
        ged.hugo_gene_symbol,
        COUNT(DISTINCT ged.sample_unique_id) as profiled_samples
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND st.sample_type IN ('Primary', 'Metastasis')
    GROUP BY st.sample_type, ged.hugo_gene_symbol
)
SELECT 
    gmbt.sample_type,
    gmbt.hugo_gene_symbol,
    gmbt.altered_samples,
    ps.profiled_samples,
    ROUND(gmbt.altered_samples * 100.0 / NULLIF(ps.profiled_samples, 0), 1) as frequency_pct,
    gmbt.total_mutation_events
FROM gene_mutations_by_type gmbt
JOIN profiled_samples ps ON gmbt.sample_type = ps.sample_type 
                         AND gmbt.hugo_gene_symbol = ps.hugo_gene_symbol
WHERE ps.profiled_samples > 0
ORDER BY gmbt.sample_type, gmbt.altered_samples DESC
LIMIT 100;",
 "max_rows": 100
}

◀ result
{"rows":[{"sample_type":"Metastasis","hugo_gene_symbol":"TP53","altered_samples":4890,"profiled_samples":5026,"frequency_pct":97.3,"total_mutation_events":5091},{"sample_type":"Metastasis","hugo_gene_symbol":"KRAS","altered_samples":2326,"profiled_samples":2408,"frequency_pct":96.6,"total_mutation_events":2342},{"sample_type":"Metastasis","hugo_gene_symbol":"APC","altered_samples":1400,"profiled_samples":1462,"frequency_pct":95.8,"total_mutation_events":2001},{"sample_type":"Metastasis","hugo_gene_symbol":"PIK3CA","altered_samples":1327,"profiled_samples":1388,"frequency_pct":95.6,"total_mutation_events":1480},{"sample_type":"Metastasis","hugo_gene_symbol":"EGFR","altered_samples":779,"profiled_samples":903,"frequency_pct":86.3,"total_mutation_events":949},{"sample_type":"Metastasis","hugo_gene_symbol":"SMAD4","altered_samples":643,"profiled_samples":853,"frequency_pct":75.4,"total_mutation_events":668},{"sample_type":"Metastasis","hugo_gene_symbol":"ARID1A","altered_samples":634,"profiled_samples":686,"frequency_pct":92.4,"total_mutation_events":729},{"sample_type":"Metastasis","hugo_gene_symbol":"KMT2D","altered_samples":569,"profiled_samples":595,"frequency_pct":95.6,"total_mutation_events":691},{"sample_type":"Metastasis","hugo_gene_symbol":"KMT2C","altered_samples":552,"profiled_samples":597,"frequency_pct":92.5,"total_mutation_events":629},{"sample_type":"Metastasis","hugo_gene_symbol":"ATM","altered_samples":483,"profiled_samples":523,"frequency_pct":92.4,"total_mutation_events":554},{"sample_type":"Metastasis","hugo_gene_symbol":"CDKN2A","altered_samples":483,"profiled_samples":1335,"frequency_pct":36.2,"total_mutation_events":507},{"sample_type":"Metastasis","hugo_gene_symbol":"KEAP1","altered_samples":475,"profiled_samples":532,"frequency_pct":89.3,"total_mutation_events":496},{"sample_type":"Metastasis","hugo_gene_symbol":"GATA3","altered_samples":458,"profiled_samples":529,"frequency_pct":86.6,"total_mutation_events":477},{"sample_type":"Metastasis","hugo_gene_symbol":"STK11","altered_samples":428,"profiled_samples":506,"frequency_pct":84.6,"total_mutation_events":440},{"sample_type":"Metastasis","hugo_gene_symbol":"NF1","altered_samples":424,"profiled_samples":483,"frequency_pct":87.8,"total_mutation_events":484},{"sample_type":"Metastasis","hugo_gene_symbol":"FAT1","altered_samples":421,"profiled_samples":497,"frequency_pct":84.7,"total_mutation_events":474},{"sample_type":"Metastasis","hugo_gene_symbol":"PTEN","altered_samples":407,"profiled_samples":697,"frequency_pct":58.4,"total_mutation_events":435},{"sample_type":"Metastasis","hugo_gene_symbol":"CDH1","altered_samples":406,"profiled_samples":457,"frequency_pct":88.8,"total_mutation_events":417},{"sample_type":"Metastasis","hugo_gene_symbol":"PTPRT","altered_samples":401,"profiled_samples":506,"frequency_pct":79.2,"total_mutation_events":453},{"sample_type":"Metastasis","hugo_gene_symbol":"ZFHX3","altered_samples":382,"profiled_samples":463,"frequency_pct":82.5,"total_mutation_ … (11982 more chars)

▶ clickhouse_run_select_query
{
 "query": "-- Compare tumor mutational burden (TMB) between Primary and Metastatic samples
WITH sample_types AS (
    SELECT 
        sample_unique_id,
        MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type,
        MAX(CASE WHEN attribute_name = 'TMB_NONSYNONYMOUS' THEN CAST(attribute_value AS Float64) END) as tmb
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name IN ('SAMPLE_TYPE', 'TMB_NONSYNONYMOUS')
    GROUP BY sample_unique_id
    HAVING sample_type IS NOT NULL AND tmb IS NOT NULL
)
SELECT
    sample_type,
    COUNT(*) as samples,
    ROUND(AVG(tmb), 2) as avg_tmb,
    ROUND(MIN(tmb), 2) as min_tmb,
    ROUND(MAX(tmb), 2) as max_tmb,
    ROUND(quantile(0.5)(tmb), 2) as median_tmb
FROM sample_types
GROUP BY sample_type
ORDER BY sample_type;",
 "max_rows": 100
}

◀ result
{"error_message":"Query execution failed: HTTPDriver for https://dl96orhu96.us-east-1.aws.clickhouse.cloud:8443 received ClickHouse error code 6\n Code: 6. DB::Exception: Cannot parse string 'Primary' as Float64: syntax error at begin of string. Note: there are toFloat64OrZero and toFloat64OrNull functions, which returns zero/NULL instead of throwing exception: while executing 'FUNCTION CAST(__table3.attribute_value :: 0, 'Float64'_String :: 5) -> CAST(__table3.attribute_value, 'Float64'_String) Float64 : 4'. (CANNOT_PARSE_TEXT) (version 26.4.1.2359 (official build))\n"}

▶ clickhouse_run_select_query
{
 "query": "-- Top 20 mutated genes in PRIMARY samples only
WITH sample_types AS (
    SELECT sample_unique_id,
           MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name = 'SAMPLE_TYPE'
    GROUP BY sample_unique_id
),
gene_mutations AS (
    SELECT
        'Primary' as sample_type,
        ged.hugo_gene_symbol,
        COUNT(DISTINCT ged.sample_unique_id) as altered_samples,
        COUNT(*) as total_mutation_events
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND ged.variant_type = 'mutation'
      AND ged.mutation_status != 'UNCALLED'
      AND ged.off_panel = 0
      AND st.sample_type = 'Primary'
    GROUP BY ged.hugo_gene_symbol
),
profiled_samples AS (
    SELECT 
        ged.hugo_gene_symbol,
        COUNT(DISTINCT ged.sample_unique_id) as profiled_samples
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND st.sample_type = 'Primary'
    GROUP BY ged.hugo_gene_symbol
)
SELECT 
    gm.hugo_gene_symbol,
    gm.altered_samples,
    ps.profiled_samples,
    ROUND(gm.altered_samples * 100.0 / NULLIF(ps.profiled_samples, 0), 1) as frequency_pct,
    gm.total_mutation_events
FROM gene_mutations gm
JOIN profiled_samples ps ON gm.hugo_gene_symbol = ps.hugo_gene_symbol
WHERE ps.profiled_samples > 0
ORDER BY gm.altered_samples DESC
LIMIT 20;",
 "max_rows": 100
}

◀ result
{"rows":[{"hugo_gene_symbol":"TP53","altered_samples":8104,"profiled_samples":8219,"frequency_pct":98.6,"total_mutation_events":8653},{"hugo_gene_symbol":"KRAS","altered_samples":4751,"profiled_samples":4848,"frequency_pct":98,"total_mutation_events":4838},{"hugo_gene_symbol":"APC","altered_samples":3345,"profiled_samples":3388,"frequency_pct":98.7,"total_mutation_events":5144},{"hugo_gene_symbol":"PIK3CA","altered_samples":2351,"profiled_samples":2442,"frequency_pct":96.3,"total_mutation_events":2665},{"hugo_gene_symbol":"EGFR","altered_samples":1338,"profiled_samples":1502,"frequency_pct":89.1,"total_mutation_events":1560},{"hugo_gene_symbol":"KMT2D","altered_samples":1203,"profiled_samples":1233,"frequency_pct":97.6,"total_mutation_events":1687},{"hugo_gene_symbol":"ARID1A","altered_samples":1199,"profiled_samples":1251,"frequency_pct":95.8,"total_mutation_events":1462},{"hugo_gene_symbol":"SMAD4","altered_samples":1142,"profiled_samples":1317,"frequency_pct":86.7,"total_mutation_events":1246},{"hugo_gene_symbol":"KMT2C","altered_samples":1059,"profiled_samples":1095,"frequency_pct":96.7,"total_mutation_events":1331},{"hugo_gene_symbol":"ATM","altered_samples":890,"profiled_samples":925,"frequency_pct":96.2,"total_mutation_events":1111},{"hugo_gene_symbol":"FBXW7","altered_samples":876,"profiled_samples":895,"frequency_pct":97.9,"total_mutation_events":1007},{"hugo_gene_symbol":"CDKN2A","altered_samples":846,"profiled_samples":1596,"frequency_pct":53,"total_mutation_events":896},{"hugo_gene_symbol":"ZFHX3","altered_samples":828,"profiled_samples":889,"frequency_pct":93.1,"total_mutation_events":1127},{"hugo_gene_symbol":"BRAF","altered_samples":826,"profiled_samples":888,"frequency_pct":93,"total_mutation_events":865},{"hugo_gene_symbol":"FAT1","altered_samples":818,"profiled_samples":874,"frequency_pct":93.6,"total_mutation_events":1118},{"hugo_gene_symbol":"PTEN","altered_samples":782,"profiled_samples":1063,"frequency_pct":73.6,"total_mutation_events":946},{"hugo_gene_symbol":"PTPRT","altered_samples":780,"profiled_samples":918,"frequency_pct":85,"total_mutation_events":926},{"hugo_gene_symbol":"NF1","altered_samples":778,"profiled_samples":851,"frequency_pct":91.4,"total_mutation_events":962},{"hugo_gene_symbol":"STK11","altered_samples":755,"profiled_samples":827,"frequency_pct":91.3,"total_mutation_events":793},{"hugo_gene_symbol":"RBM10","altered_samples":732,"profiled_samples":778,"frequency_pct":94.1,"total_mutation_events":759}]}

▶ clickhouse_run_select_query
{
 "query": "-- Compare mutation frequencies for top genes: Primary vs Metastasis
SELECT 
    hugo_gene_symbol,
    MAX(CASE WHEN sample_type = 'Primary' THEN altered_samples END) as primary_altered,
    MAX(CASE WHEN sample_type = 'Primary' THEN frequency_pct END) as primary_pct,
    MAX(CASE WHEN sample_type = 'Metastasis' THEN altered_samples END) as met_altered,
    MAX(CASE WHEN sample_type = 'Metastasis' THEN frequency_pct END) as met_pct,
    ROUND(MAX(CASE WHEN sample_type = 'Metastasis' THEN frequency_pct END) - 
          MAX(CASE WHEN sample_type = 'Primary' THEN frequency_pct END), 1) as pct_difference
FROM (
    WITH sample_types AS (
        SELECT sample_unique_id,
               MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
        FROM clinical_data_derived
        WHERE cancer_study_identifier = 'msk_chord_2024'
          AND attribute_name = 'SAMPLE_TYPE'
        GROUP BY sample_unique_id
    ),
    gene_mutations_by_type AS (
        SELECT
            st.sample_type,
            ged.hugo_gene_symbol,
            COUNT(DISTINCT ged.sample_unique_id) as altered_samples,
            COUNT(*) as total_mutation_events
        FROM genomic_event_derived ged
        JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
        WHERE ged.cancer_study_identifier = 'msk_chord_2024'
          AND ged.variant_type = 'mutation'
          AND ged.mutation_status != 'UNCALLED'
          AND ged.off_panel = 0
          AND st.sample_type IN ('Primary', 'Metastasis')
        GROUP BY st.sample_type, ged.hugo_gene_symbol
    ),
    profiled_samples AS (
        SELECT 
            st.sample_type,
            ged.hugo_gene_symbol,
            COUNT(DISTINCT ged.sample_unique_id) as profiled_samples
        FROM genomic_event_derived ged
        JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
        WHERE ged.cancer_study_identifier = 'msk_chord_2024'
          AND st.sample_type IN ('Primary', 'Metastasis')
        GROUP BY st.sample_type, ged.hugo_gene_symbol
    )
    SELECT 
        gmbt.sample_type,
        gmbt.hugo_gene_symbol,
        gmbt.altered_samples,
        ps.profiled_samples,
        ROUND(gmbt.altered_samples * 100.0 / NULLIF(ps.profiled_samples, 0), 1) as frequency_pct
    FROM gene_mutations_by_type gmbt
    JOIN profiled_samples ps ON gmbt.sample_type = ps.sample_type 
                             AND gmbt.hugo_gene_symbol = ps.hugo_gene_symbol
    WHERE ps.profiled_samples > 0
)
WHERE hugo_gene_symbol IN (
    SELECT hugo_gene_symbol FROM (
        WITH sample_types AS (
            SELECT sample_unique_id,
                   MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
            FROM clinical_data_derived
            WHERE cancer_study_identifier = 'msk_chord_2024' AND attribute_name = 'SAMPLE_TYPE'
            GROUP BY sample_unique_id
        )
        SELECT 
            ged.hugo_gene_symbol,
            COUNT(DISTINCT ged.sample_unique_id) as altered_samples
        FROM genomic_event_derived ged
        JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
        WHERE ged.cancer_study_identifier = 'msk_chord_2024'
          AND ged.variant_type = 'mutation'
          AND ged.mutation_status != 'UNCALLED'
          AND ged.off_panel = 0
          AND st.sample_type = 'Primary'
        GROUP BY ged.hugo_gene_symbol
        ORDER BY altered_samples DESC
        LIMIT 20
    )
)
GROUP BY hugo_gene_symbol
ORDER BY primary_pct DESC, hugo_gene_symbol;",
 "max_rows": 100
}

◀ result
{"rows":[{"hugo_gene_symbol":"APC","primary_altered":3345,"primary_pct":98.7,"met_altered":1400,"met_pct":95.8,"pct_difference":-2.9},{"hugo_gene_symbol":"TP53","primary_altered":8104,"primary_pct":98.6,"met_altered":4890,"met_pct":97.3,"pct_difference":-1.3},{"hugo_gene_symbol":"KRAS","primary_altered":4751,"primary_pct":98,"met_altered":2326,"met_pct":96.6,"pct_difference":-1.4},{"hugo_gene_symbol":"FBXW7","primary_altered":876,"primary_pct":97.9,"met_altered":238,"met_pct":91.9,"pct_difference":-6},{"hugo_gene_symbol":"KMT2D","primary_altered":1203,"primary_pct":97.6,"met_altered":569,"met_pct":95.6,"pct_difference":-2},{"hugo_gene_symbol":"KMT2C","primary_altered":1059,"primary_pct":96.7,"met_altered":552,"met_pct":92.5,"pct_difference":-4.2},{"hugo_gene_symbol":"PIK3CA","primary_altered":2351,"primary_pct":96.3,"met_altered":1327,"met_pct":95.6,"pct_difference":-0.7},{"hugo_gene_symbol":"ATM","primary_altered":890,"primary_pct":96.2,"met_altered":483,"met_pct":92.4,"pct_difference":-3.8},{"hugo_gene_symbol":"ARID1A","primary_altered":1199,"primary_pct":95.8,"met_altered":634,"met_pct":92.4,"pct_difference":-3.4},{"hugo_gene_symbol":"RBM10","primary_altered":732,"primary_pct":94.1,"met_altered":323,"met_pct":82.8,"pct_difference":-11.3},{"hugo_gene_symbol":"FAT1","primary_altered":818,"primary_pct":93.6,"met_altered":421,"met_pct":84.7,"pct_difference":-8.9},{"hugo_gene_symbol":"ZFHX3","primary_altered":828,"primary_pct":93.1,"met_altered":382,"met_pct":82.5,"pct_difference":-10.6},{"hugo_gene_symbol":"BRAF","primary_altered":826,"primary_pct":93,"met_altered":307,"met_pct":86,"pct_difference":-7},{"hugo_gene_symbol":"NF1","primary_altered":778,"primary_pct":91.4,"met_altered":424,"met_pct":87.8,"pct_difference":-3.6},{"hugo_gene_symbol":"STK11","primary_altered":755,"primary_pct":91.3,"met_altered":428,"met_pct":84.6,"pct_difference":-6.7},{"hugo_gene_symbol":"EGFR","primary_altered":1338,"primary_pct":89.1,"met_altered":779,"met_pct":86.3,"pct_difference":-2.8},{"hugo_gene_symbol":"SMAD4","primary_altered":1142,"primary_pct":86.7,"met_altered":643,"met_pct":75.4,"pct_difference":-11.3},{"hugo_gene_symbol":"PTPRT","primary_altered":780,"primary_pct":85,"met_altered":401,"met_pct":79.2,"pct_difference":-5.8},{"hugo_gene_symbol":"PTEN","primary_altered":782,"primary_pct":73.6,"met_altered":407,"met_pct":58.4,"pct_difference":-15.2},{"hugo_gene_symbol":"CDKN2A","primary_altered":846,"primary_pct":53,"met_altered":483,"met_pct":36.2,"pct_difference":-16.8}]}

▶ clickhouse_run_select_query
{
 "query": "-- Compare TMB between Primary and Metastatic samples (fixed)
WITH tmb_data AS (
    SELECT 
        sample_unique_id,
        MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type,
        MAX(CASE WHEN attribute_name = 'TMB_NONSYNONYMOUS' AND attribute_value NOT IN ('', 'NA') 
                 THEN CAST(attribute_value AS Float64) END) as tmb
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name IN ('SAMPLE_TYPE', 'TMB_NONSYNONYMOUS')
    GROUP BY sample_unique_id
)
SELECT
    sample_type,
    COUNT(*) as samples_with_tmb,
    ROUND(AVG(tmb), 2) as avg_tmb,
    ROUND(MIN(tmb), 2) as min_tmb,
    ROUND(MAX(tmb), 2) as max_tmb,
    ROUND(quantile(0.5)(tmb), 2) as median_tmb,
    ROUND(quantile(0.25)(tmb), 2) as q1_tmb,
    ROUND(quantile(0.75)(tmb), 2) as q3_tmb
FROM tmb_data
WHERE sample_type IS NOT NULL AND tmb IS NOT NULL
GROUP BY sample_type
ORDER BY sample_type;",
 "max_rows": 100
}

◀ result
{"error_message":"Query execution failed: HTTPDriver for https://dl96orhu96.us-east-1.aws.clickhouse.cloud:8443 received ClickHouse error code 6\n Code: 6. DB::Exception: Cannot parse string 'Primary' as Float64: syntax error at begin of string. Note: there are toFloat64OrZero and toFloat64OrNull functions, which returns zero/NULL instead of throwing exception: while executing 'FUNCTION CAST(__table3.attribute_value : 0, 'Float64'_String :: 5) -> CAST(__table3.attribute_value, 'Float64'_String) Float64 : 4'. (CANNOT_PARSE_TEXT) (version 26.4.1.2359 (official build))\n"}

▶ clickhouse_run_select_query
{
 "query": "-- Compare copy-number alterations (CNA): Primary vs Metastasis
WITH sample_types AS (
    SELECT sample_unique_id,
           MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name = 'SAMPLE_TYPE'
    GROUP BY sample_unique_id
),
cna_counts AS (
    SELECT
        st.sample_type,
        ged.hugo_gene_symbol,
        CASE WHEN ged.cna_alteration = 2 THEN 'Amplification'
             WHEN ged.cna_alteration = -2 THEN 'Deep Deletion'
        END as cna_type,
        COUNT(DISTINCT ged.sample_unique_id) as altered_samples,
        COUNT(*) as total_cna_events
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND ged.variant_type = 'cna'
      AND st.sample_type IN ('Primary', 'Metastasis')
      AND ged.cna_alteration IN (2, -2)
    GROUP BY st.sample_type, ged.hugo_gene_symbol, ged.cna_alteration
),
cna_profiled AS (
    SELECT 
        st.sample_type,
        ged.hugo_gene_symbol,
        COUNT(DISTINCT ged.sample_unique_id) as profiled_samples
    FROM genomic_event_derived ged
    JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
    WHERE ged.cancer_study_identifier = 'msk_chord_2024'
      AND st.sample_type IN ('Primary', 'Metastasis')
    GROUP BY st.sample_type, ged.hugo_gene_symbol
)
SELECT 
    cc.sample_type,
    cc.hugo_gene_symbol,
    cc.cna_type,
    cc.altered_samples,
    cp.profiled_samples,
    ROUND(cc.altered_samples * 100.0 / NULLIF(cp.profiled_samples, 0), 1) as frequency_pct
FROM cna_counts cc
JOIN cna_profiled cp ON cc.sample_type = cp.sample_type 
                     AND cc.hugo_gene_symbol = cp.hugo_gene_symbol
WHERE cp.profiled_samples > 0
ORDER BY cc.sample_type, cc.altered_samples DESC
LIMIT 40;",
 "max_rows": 100
}

◀ result
{"rows":[{"sample_type":"Metastasis","hugo_gene_symbol":"CDKN2A","cna_type":"Deep Deletion","altered_samples":826,"profiled_samples":1335,"frequency_pct":61.9},{"sample_type":"Metastasis","hugo_gene_symbol":"MYC","cna_type":"Amplification","altered_samples":772,"profiled_samples":825,"frequency_pct":93.6},{"sample_type":"Metastasis","hugo_gene_symbol":"CDKN2B","cna_type":"Deep Deletion","altered_samples":770,"profiled_samples":807,"frequency_pct":95.4},{"sample_type":"Metastasis","hugo_gene_symbol":"CCND1","cna_type":"Amplification","altered_samples":663,"profiled_samples":689,"frequency_pct":96.2},{"sample_type":"Metastasis","hugo_gene_symbol":"FGF19","cna_type":"Amplification","altered_samples":617,"profiled_samples":649,"frequency_pct":95.1},{"sample_type":"Metastasis","hugo_gene_symbol":"FGF4","cna_type":"Amplification","altered_samples":600,"profiled_samples":648,"frequency_pct":92.6},{"sample_type":"Metastasis","hugo_gene_symbol":"FGF3","cna_type":"Amplification","altered_samples":582,"profiled_samples":659,"frequency_pct":88.3},{"sample_type":"Metastasis","hugo_gene_symbol":"FGFR1","cna_type":"Amplification","altered_samples":524,"profiled_samples":742,"frequency_pct":70.6},{"sample_type":"Metastasis","hugo_gene_symbol":"ERBB2","cna_type":"Amplification","altered_samples":425,"profiled_samples":655,"frequency_pct":64.9},{"sample_type":"Metastasis","hugo_gene_symbol":"RECQL4","cna_type":"Amplification","altered_samples":422,"profiled_samples":562,"frequency_pct":75.1},{"sample_type":"Metastasis","hugo_gene_symbol":"NSD3","cna_type":"Amplification","altered_samples":386,"profiled_samples":514,"frequency_pct":75.1},{"sample_type":"Metastasis","hugo_gene_symbol":"RAD21","cna_type":"Amplification","altered_samples":353,"profiled_samples":414,"frequency_pct":85.3},{"sample_type":"Metastasis","hugo_gene_symbol":"AGO2","cna_type":"Amplification","altered_samples":311,"profiled_samples":396,"frequency_pct":78.5},{"sample_type":"Metastasis","hugo_gene_symbol":"CDK12","cna_type":"Amplification","altered_samples":299,"profiled_samples":485,"frequency_pct":61.6},{"sample_type":"Metastasis","hugo_gene_symbol":"EGFR","cna_type":"Amplification","altered_samples":289,"profiled_samples":903,"frequency_pct":32},{"sample_type":"Metastasis","hugo_gene_symbol":"PTEN","cna_type":"Deep Deletion","altered_samples":283,"profiled_samples":697,"frequency_pct":40.6},{"sample_type":"Metastasis","hugo_gene_symbol":"PAK1","cna_type":"Amplification","altered_samples":267,"profiled_samples":315,"frequency_pct":84.8},{"sample_type":"Metastasis","hugo_gene_symbol":"AR","cna_type":"Amplification","altered_samples":265,"profiled_samples":551,"frequency_pct":48.1},{"sample_type":"Metastasis","hugo_gene_symbol":"MDM2","cna_type":"Amplification","altered_samples":263,"profiled_samples":312,"frequency_pct":84.3},{"sample_type":"Metastasis","hugo_gene_symbol":"NKX2-1","cna_type":"Amplification","altered_samples":260,"profiled_samples":321,"frequency_pct":81},{"sample_type":"Metasta … (2942 more chars)

▶ clickhouse_run_select_query
{
 "query": "-- Structural variants in Primary vs Metastasis samples
WITH sample_types AS (
    SELECT sample_unique_id,
           MAX(CASE WHEN attribute_name = 'SAMPLE_TYPE' THEN attribute_value END) as sample_type
    FROM clinical_data_derived
    WHERE cancer_study_identifier = 'msk_chord_2024'
      AND attribute_name = 'SAMPLE_TYPE'
    GROUP BY sample_unique_id
)
SELECT
    st.sample_type,
    COUNT(DISTINCT ged.sample_unique_id) as samples_with_svs,
    COUNT(DISTINCT ged.hugo_gene_symbol) as distinct_genes_with_sv,
    COUNT(*) as total_sv_events
FROM genomic_event_derived ged
JOIN sample_types st ON ged.sample_unique_id = st.sample_unique_id
WHERE ged.cancer_study_identifier = 'msk_chord_2024'
  AND ged.variant_type = 'structural_variant'
  AND st.sample_type IN ('Primary', 'Metastasis')
GROUP BY st.sample_type
ORDER BY st.sample_type;",
 "max_rows": 100
}

◀ result
{"rows":[{"sample_type":"Metastasis","samples_with_svs":2196,"distinct_genes_with_sv":1179,"total_sv_events":4239},{"sample_type":"Primary","samples_with_svs":3133,"distinct_genes_with_sv":1430,"total_sv_events":5910}]}

▶ resolve_and_route
{
 "studyIds": [
  "msk_chord_2024"
 ]
}

◀ result
{"success":true,"message":"Found 1 study. Use the metadata to call the appropriate navigation tool(s).","data":{"totalCount":1,"studiesWithMetadata":[{"studyId":"msk_chord_2024","name":"MSK-CHORD (MSK, Nature 2024)","sampleCount":25040,"studyViewUrl":"https://www.cbioportal.org/study?id=msk_chord_2024","metadata":{"clinicalAttributeIds":["ADRENAL_GLANDS","BONE","CANCER_TYPE","CANCER_TYPE_DETAILED","CLINICAL_GROUP","CLINICAL_SUMMARY","CNS_BRAIN","CURRENT_AGE_DEID","DIAGNOSIS_DESCRIPTION","ETHNICITY","FRACTION_GENOME_ALTERED","GENDER","GENE_PANEL","GLEASON_FIRST_REPORTED","GLEASON_HIGHEST_REPORTED","GLEASON_SAMPLE_LEVEL","HER2","HISTORY_OF_PDL1","HR","ICD_O_HISTOLOGY_DESCRIPTION","INTRA_ABDOMINAL","LIVER","LUNG","LYMPH_NODES","METASTATIC_SITE","MSI_COMMENT","MSI_SCORE","MSI_TYPE","MUTATION_COUNT","NUM_ICDO_DX","ONCOTREE_CODE","OS_MONTHS","OS_STATUS","OTHER","PATHOLOGICAL_GROUP","PDL1_POSITIVE","PLEURA","PRIMARY_SITE","PRIOR_MED_TO_MSK","RACE","REPRODUCTIVE_ORGANS","SAMPLE_CLASS","SAMPLE_COUNT","SAMPLE_COVERAGE","SAMPLE_TYPE","SMOKING_PREDICTIONS_3_CLASSES","SOMATIC_STATUS","STAGE_HIGHEST_RECORDED","TMB_NONSYNONYMOUS","TUMOR_PURITY"],"molecularProfileIds":["msk_chord_2024_cna","msk_chord_2024_mutations","msk_chord_2024_structural_variants"],"availableComparisonTabs":["overlap","clinical","survival","alterations","mutations"]}}]}}

▶ navigate_to_study_view
{
 "studyIds": [
  "msk_chord_2024"
 ],
 "filterJson": {
  "clinicalDataFilters": [
   {
    "attributeId": "SAMPLE_TYPE",
    "values": [
     {
      "value": "Primary"
     },
     {
      "value": "Metastasis"
     }
    ]
   }
  ]
 }
}

◀ result
{"success":true,"message":"Navigating to https://www.cbioportal.org/study/summary?id=msk_chord_2024#filterJson=%7B%22clinicalDataFilters%22%3A%5B%7B%22attributeId%22%3A%22SAMPLE_TYPE%22%2C%22values%22%3A%5B%7B%22value%22%3A%22Primary%22%7D%2C%7B%22value%22%3A%22Metastasis%22%7D%5D%7D%5D%7D","url":"https://www.cbioportal.org/study/summary?id=msk_chord_2024#filterJson=%7B%22clinicalDataFilters%22%3A%5B%7B%22attributeId%22%3A%22SAMPLE_TYPE%22%2C%22values%22%3A%5B%7B%22value%22%3A%22Primary%22%7D%2C%7B%22value%22%3A%22Metastasis%22%7D%5D%7D%5D%7D","data":{"studyIds":["msk_chord_2024"],"studies":[{"studyId":"msk_chord_2024","name":"MSK-CHORD (MSK, Nature 2024)","sampleCount":25040}],"tab":"summary","hasFilters":true,"hasPlotsConfig":false,"pageDescription":"Overview of all study data displayed as interactive charts summarizing available clinical and genomic data. Click within a chart to filter. Additional charts may be available in the Charts menu."}}

▶ navigate_to_group_comparison
{
 "studyIds": [
  "msk_chord_2024"
 ],
 "groups": [
  {
   "name": "Primary",
   "studyViewFilter": {
    "clinicalDataFilters": [
     {
      "attributeId": "SAMPLE_TYPE",
      "values": [
       {
        "value": "Primary"
       }
      ]
     }
    ]
   }
  },
  {
   "name": "Metastasis",
   "studyViewFilter": {
    "clinicalDataFilters": [
     {
      "attributeId": "SAMPLE_TYPE",
      "values": [
       {
        "value": "Metastasis"
       }
      ]
     }
    ]
   }
  }
 ],
 "tab": "overlap"
}

◀ result
{"success":true,"message":"Navigating to https://www.cbioportal.org/comparison/overlap?comparisonId=6ab49e90c2115c492d884a5c","url":"https://www.cbioportal.org/comparison/overlap?comparisonId=6ab49e90c2115c492d884a5c","data":{"description":"Group comparison (2 custom groups)","studies":["msk_chord_2024"],"totalGroups":2,"groups":[{"name":"Primary","sampleCount":15928},{"name":"Metastasis","sampleCount":8878}],"studyViewUrl":"https://www.cbioportal.org/study?id=msk_chord_2024","groupUrls":[{"groupName":"Primary","url":"https://www.cbioportal.org/study?id=msk_chord_2024#filterJson=%7B%22clinicalDataFilters%22%3A%5B%7B%22attributeId%22%3A%22SAMPLE_TYPE%22%2C%22values%22%3A%5B%7B%22value%22%3A%22Primary%22%7D%5D%7D%5D%7D"},{"groupName":"Metastasis","url":"https://www.cbioportal.org/study?id=msk_chord_2024#filterJson=%7B%22clinicalDataFilters%22%3A%5B%7B%22attributeId%22%3A%22SAMPLE_TYPE%22%2C%22values%22%3A%5B%7B%22value%22%3A%22Metastasis%22%7D%5D%7D%5D%7D"}],"pageDescription":"Diagram showing sample and patient overlap across Primary (15928) vs Metastasis (8878)."}}

═ answer (success)
## Mutation Landscape: Primary vs Metastatic Samples in MSK-CHORD

### ⚠️ **Important Study Design Limitation**

The MSK-CHORD study lacks paired primary-metastatic samples from the same patients:
- **24,950 patients / 25,040 samples** (nearly 1:1 ratio)
- Only **26 patients** have BOTH a Primary AND Metastasis sample

**This analysis presents an unpaired comparison** of all Primary vs Metastasis samples across the cohort, not longitudinal paired evolution.

---

### **Sample Type Distribution**

| Sample Type | Count | Patients |
|---|---|---|
| **Primary** | 15,928 | 15,872 |
| **Metastasis** | 8,878 | 8,872 |
| Unknown | 136 | 136 |
| Local Recurrence | 98 | 98 |

---

### **Key Mutation Frequency Differences (Top 20 Genes)**

Metastatic samples show **systematically lower mutation frequencies** across top cancer drivers:

| Gene | Primary % | Metastasis % | Difference |
|---|---|---|---|
| **CDKN2A** | 53.0% | 36.2% | **−16.8%** |
| **PTEN** | 73.6% | 58.4% | **−15.2%** |
| **SMAD4** | 86.7% | 75.4% | **−11.3%** |
| **RBM10** | 94.1% | 82.8% | **−11.3%** |
| **ZFHX3** | 93.1% | 82.5% | **−10.6%** |
| **FAT1** | 93.6% | 84.7% | **−8.9%** |
| **BRAF** | 93.0% | 86.0% | **−7.0%** |
| **STK11** | 91.3% | 84.6% | **−6.7%** |
| **FBXW7** | 97.9% | 91.9% | **−6.0%** |
| **PTPRT** | 85.0% | 79.2% | **−5.8%** |
| ARID1A | 95.8% | 92.4% | −3.4% |
| ATM | 96.2% | 92.4% | −3.8% |
| KMT2D | 97.6% | 95.6% | −2.0% |
| EGFR | 89.1% | 86.3% | −2.8% |
| KRAS | 98.0% | 96.6% | −1.4% |
| TP53 | 98.6% | 97.3% | −1.3% |
| PIK3CA | 96.3% | 95.6% | −0.7% |
| APC | 98.7% | 95.8% | **−2.9%** |
| KMT2C | 96.7% | 92.5% | −4.2% |

**Interpretation**: The most dramatic losses occur in **tumor suppressors** (CDKN2A, PTEN, SMAD4, RBM10) — suggesting that **metastatic samples have enriched clones with selective loss or inactivation** of these genes. Alternatively, **sampling bias**: metastases are often small biopsies that may undersample certain mutations if they're spatially heterogeneous.

---

### **Copy-Number Alteration (CNA) Comparison**

**Top Amplifications in Metastases:**
- MYC: 772/825 samples (93.6%)
- CCND1: 663/689 (96.2%)
- FGF19, FGF4, FGF3: ~88–95%
- ERBB2: 425/655 (64.9%)

**Deep Deletions in Metastases:**
- CDKN2A: 826/1,335 (61.9%)
- CDKN2B: 770/807 (95.4%)
- NKX3-1: 260/292 (89%)
- MTAP: 188/196 (95.9%)

**Key Finding**: Metastatic samples show **very high CNA frequencies overall** — consistent with genomic instability and clonal selection during metastatic progression.

---

### **Structural Variant (SV) Burden**

| Sample Type | Samples with SVs | Distinct Genes | Total SV Events |
|---|---|---|---|
| **Primary** | 3,133 | 1,430 | 5,910 |
| **Metastasis** | 2,196 | 1,179 | 4,239 |

**Metastases have ~30% fewer SV-bearing samples** — may reflect sampling effects (smaller metastatic biopsies) or clonal selection for chromosomally stable populations during dissemination.

---

### **Summary**

1. **Tumor suppressors are depleted in metastases** (CDKN2A −16.8%, PTEN −15.2%, SMAD4 −11.3%), suggesting either:
   - Metastases evolved from clones that lost these genes early
   - Sampling bias from small biopsy cores

2. **Core driver mutations (TP53, KRAS, APC) are retained at similar frequencies**, indicating early truncal origin

3. **CNA burden is high and relatively uniform** across both groups, suggesting chromosomal instability as a general feature

4. **SV burden is lower in metastases**, possibly due to smaller sample sizes (biopsies vs resections)

---

### **Interactive Analysis**

[View Primary vs Metastasis cohort overview](https://www.cbioportal.org/study/summary?id=msk_chord_2024#filterJson=%7B%22clinicalDataFilters%22%3A%5B%7B%22attributeId%22%3A%22SAMPLE_TYPE%22%2C%22values%22%3A%5B%7B%22value%22%3A%22Primary%22%7D%2C%7B%22value%22%3A%22Metastasis%22%7D%5D%7D%5D%7D)

[Compare groups side-by-side (mutations, survival, clinical traits)](https://www.cbioportal.org/comparison/overlap?comparisonId=6ab49e90c2115c492d884a5c)