Skip to lesson
OncoGuideeducationDiana’s wiki
THE EDUCATION LIBRARY

ctDNA PPM and MRI reads

ctDNA and MRI answer different questions. A blood test asks, "Did this plasma tube contain tumor-matching DNA fragments above the assay's detection floor?" MRI asks, "What tissue still enhances, shrinks, persists, or looks structurally abnormal?"

Those questions overlap, but they are not the same measurement. The hardest part is that both can be true at once: a tumor can be present but not shedding enough ctDNA to detect, and ctDNA can be detectable when MRI cannot localize anything.

For Diana's current moment, the useful frame is:

  • Personalis 304 PPM -> 1 PPM means the shedding signal collapsed.
  • 1 PPM detected means molecular response, not molecular clearance.
  • 0 PPM / not detected would mean below detection in that draw, not proof that every tumor cell is gone.
  • MRI will translate that molecular response into an anatomic read: how much breast enhancement, nodal abnormality, and treatment-effect signal remains.

This page is an educational mental model. Diana's live result trend stays in results and the operational MRD evidence stack stays in ctdna-mrd.

1 PPM is a sparse blood-sample signal

ppm blood snapshot cartoon

1 PPM means the assay found tumor-matching fragments at roughly one part per million cell-free DNA molecules. It does not mean one tumor cell was floating in the blood.

The tube contains mostly normal cell-free DNA. The tumor-derived fragments are the rare marked pieces in that background. At very low levels, the qualitative call carries more weight than the exact number: detected versus not detected is sturdier than arguing whether 1 PPM is exactly half or twice some nearby value.

That is why Personalis' own low-PPM caution matters for Diana's June 9 result. 1 PPM should be read as very-low-positive, not a precise body-wide tumor-volume estimate.

0 PPM is not the same as zero tumor cells

no shedding residual tumor cartoon

0 PPM or not detected means the assay did not find enough tumor-derived DNA in that blood draw. It does not prove there is no residual tumor anywhere.

A tumor can be present but under-shed into plasma for several reasons:

  • Tiny remaining volume: too few cells are turning over to cross the detection floor.
  • Local compartment: a residual breast or nodal focus may not release much DNA into systemic blood.
  • Quiet biology: treatment can leave lower-turnover viable cells or dormant microscopic disease.
  • Sampling limits: the tube captures a finite slice of a fast-moving blood compartment.
  • Clearance is fast: ctDNA fragments are short-lived, so a draw reflects recent shedding, not a permanent archive.

This is the clean way to say the paradox: ctDNA negativity is reassuring, but it is not omniscient.

MRI sees anatomy; ctDNA sees shedding

mri ctdna discordance cartoon

MRI and ctDNA can land in four broad patterns:

PatternWhat it can mean
MRI smaller + ctDNA falling/clearingBest-aligned response pattern. Tumor burden is falling by both anatomy and shedding.
MRI abnormal + ctDNA low/negativeCould be residual tumor, but also treatment effect, fibrosis, necrosis, edema, or non-mass enhancement.
MRI clean + ctDNA positiveCould be microscopic residual disease below imaging resolution or outside the MRI field.
MRI abnormal + ctDNA positiveHigher concern for viable residual disease, especially if both are persistent at later gates.

MRI is especially tricky after neoadjuvant therapy because enhancement is not pure tumor. Treated tissue can still enhance, and non-mass enhancement can overstate residual viable disease. The opposite is also possible: microscopic tumor can be below MRI resolution while ctDNA still detects molecular activity.

What this means for Friday's MRI

Diana's Personalis result makes a strong anatomic response more likely than stable or progressive disease. The current result is not "molecularly clean," though, so a Friday MRI that still shows residual enhancement or residual nodal abnormality would not contradict the blood trend.

The most coherent expectation is:

  • Marked response would fit the 304 -> 1 PPM collapse.
  • Near-complete MRI response would be excellent, but still not the same as pCR or MRD clearance.
  • Residual MRI abnormality would remain plausible because MRI sees anatomy and treatment effect, not just living tumor.
  • Clear progression would be more surprising and would deserve careful reconciliation against the molecular response.

The final risk interpretation will come from the combined matrix: MRI response now, late-June / post-NAT ctDNA status, and surgical pathology / RCB later.