Consistency of clinical data from TCGA (melanoma)

Question

[rachelpachel]

I'm trying to understand the clinical data for the TCGA melanoma (nationwidechildrens.org_clinical_patient_skcm.txt). From sources like https://cancerstaging.org/references-tools/quickreferences/Documents/MelanomaMedium.pdf, stage 0 (melanoma in situ, also graded Tis) is inherently not metastatic and that is why it is labeled stage 0.

But in the clinical data (partially copied below), samples that are labeled Stage 0 / Tis with no regional or distant metastases (N0 and M0), are also labeled as distant metastases. Also the patients are dead, presumably of their metastatic melanomas. This seems impossible. Maybe it was initially diagnosed as stage 0 and somehow became metastatic anyway? --but then you would not expect that to be the annotation of that tumor data in the clinical file. If I want to find all of the clinically low stage tumors, what would be the best way to do that, if not by using this field? Any ideas? Thank you!

bcr_patient_barcode    tumor_status    vital_status    death_days_to    primary_melanoma_known_dx    primary_multiple_at_dx    primary_at_dx_count    breslow_thickness_at_diagnosis    clark_level_at_diagnosis    primary_melanoma_tumor_ulceration    ajcc_tumor_pathologic_pt    ajcc_nodes_pathologic_pn    ajcc_metastasis_pathologic_pm    ldh_level          ajcc_pathologic_tumor_stage    submitted_tumor_site    submitted_tumor_site    metastatic_tumor_site    primary_melanoma_skin_type
TCGA-EE-A20C           WITH TUMOR      Dead            4601             YES                          NO                        [Not Available]        [Not Available]                   I                           [Unknown]                            Tis                         N0                          M0                               [Not Available]    Stage 0                        Extremities             Distant Metastasis      Right back               Non-glabrous skin
TCGA-ER-A2NE           WITH TUMOR      Dead            613              YES                          NO                        [Not Available]        [Not Available]                   [Not Available]             [Not Available]                      Tis                         N0                          M0                               [Not Available]    Stage 0                        Extremities             Distant Metastasis      Ileum (serosa)           [Not Available]
TCGA-D3-A8GR           WITH TUMOR      Dead            3943             YES                          NO                        [Not Available]        0.01                              I                           [Unknown]                            Tis                         N0                          M0                               [Not Available]    Stage 0                        Extremities             Distant Metastasis      Small Intestine          [Not Available]

Answer 1

[rachelpachel]

I think I figured it out. Emailing with the TCGA helpdesk helped a bit. There's some helpful info on [the corresponding enrollment form][1]

And most of those fields pertain to the patient's primary diagnosis. Only the fields below actually pertain to the tumor specimen (plus a few other fields having to do with therapies that would lead to exclusion of the tumor)

submitted_tumor_dx_days_to       submitted_tumor_site    metastatic_tumor_site
days_to_submitted_specimen_dx    tumor_tissue_site       distant_metastasis_anatomic_site
CDE_ID:3430945                   CDE_ID:3427536          CDE_ID:2961431

So, my interpretation of patient TCGA-ER-A2NE (from my original question), is: The patient originally presented in 2007 with a melanoma in situ with no metastasis, and that original melanoma was located on the extremities. 567 days later the patient had a metastasis in the ileum. This metastasis in the ileum is the sample that was submitted to TCGA. The patient died of the metastasis 613 days after diagnosis.

Therefore, what I was hoping to do (use the clinical data to find the lower grade/stage tumors) is only somewhat possible. Instead I used the TCGA barcodes to identify primary & metastasis. Hope this helps someone.

[1] http://www.nationwidechildrens.org/melanoma-enrollment-form

[Cyriac Kandoth]

Nice sleuthing, Rachel. And welcome to Biostars!

[rachelpachel]

heh thanks Cyriac! Nice to see you on the webs. Hope we can get a drink sometime after I finish this pesky defense :)