Metastasis or second primary cancer?

01The case

A patient has a confirmed primary pancreatic cancer. During an outpatient gastroenterology evaluation, a rectal mass is found. The specialist suspects it could be a metastasis from the pancreatic tumor.

This is where the real challenge begins: is it a metastasis, a second primary cancer, or a lesion that does not yet have a definitive diagnosis?

02Two perspectives, one responsibility

The physician has to establish the origin of the lesion through clinical judgment, diagnostic studies and, where appropriate, histopathologic evaluation.

The HCC/CDI auditor has to determine which diagnoses are documented well enough to assign ICD-10-CM codes and which need clarification. It is not the auditor's place to turn the specialist's suspicion into a confirmed diagnosis.

The ICD-10-CM Official Guidelines FY2027, Section IV.H, say that in outpatient services diagnoses documented as "probable", "suspected", "questionable", "rule out", "compatible with", "consistent with" or "working diagnosis", or other similar terms indicating uncertainty, are not coded. Instead, the condition is coded to the highest degree of certainty for that encounter: symptoms, signs, abnormal test results or other reason for the visit.

In an inpatient stay the rule is different: an uncertain diagnosis documented at discharge is coded as if it existed (Sections II.H and III.C). That is why the setting of the encounter matters.

03How does the CMS-HCC V28 model come into it?

A primary pancreatic cancer and a secondary neoplasm of the rectum are different diagnoses, and in CMS-HCC V28 they fall into different categories.

If a rectal metastasis is confirmed, the code is C78.5, while a second primary cancer of the rectum is C20. The pancreatic cancer (category C25) falls in HCC 20, lung and other severe cancers. What happens next depends on the hierarchy:

• Metastasis (C78.5): falls in HCC 17, cancer metastatic to lung, liver, brain and other organs (metastasis to bone or to an unspecified site, among others, falls in HCC 18). In the V28 hierarchy, HCC 17 sits above HCC 20 and replaces it in the score, with a much larger weight. The pancreatic cancer is still coded; the hierarchy only decides which HCC counts.
• Second primary (C20): falls in HCC 22, bladder, colorectal and other cancers. HCC 20 already sits above HCC 22 in the hierarchy, so it adds nothing to the score.

This assumes the pancreatic cancer is active or under treatment, documented in the same data collection period. If it was excised, is no longer treated and there is no evidence it is still present, it is coded as history (Z85.07), which falls in no HCC. In that case, C20 would add HCC 22.

So if the metastatic HCC is reported, this distinction is what the record has to support in a RADV validation.

04The documentation that matters in RADV

A concise, specific, clinically supported note should identify the established diagnosis, the tumor site, whether it is primary or secondary, the clinical evidence reviewed, and the plan for evaluation or treatment.

If the origin of the lesion remains uncertain, the note should say so, and the diagnostic workup continues.

No well-written note replaces the minimum requirements CMS checks in RADV, according to its most recent reviewer guidance (2020): that the record belongs to the right beneficiary; that it documents a face-to-face encounter with a provider type, source and specialty acceptable for risk adjustment; that the date of service falls within the data collection period under review; that it carries a valid signature and credentials; and that it is coded according to the official ICD-10-CM conventions, instructions and Official Guidelines.

Under the 2026 AHIMA/ACDIS guidelines, a CDI query must be neutral, grounded in clinical indicators, free of any reference to reimbursement, and must respect the physician's independent judgment. And in RADV, a query form the organization generates after the encounter is not accepted as part of the record; only a query made at or near the time of the encounter is evaluated, case by case.

The question is not how to get an additional HCC. The question is whether there is enough clinical evidence to support it.

05When the answer is in a report

A pathology report with the pathologist's interpretation is an acceptable source in RADV, as an exception to the face-to-face requirement, and a confirmed diagnosis in that interpretation can be coded. But if the report says "compatible with" or "consistent with", that is still a term of uncertainty, not a confirmed diagnosis.

A radiology report, on the other hand, is not enough on its own: the treating physician has to acknowledge the finding in their note. And if the documentation conflicts, for example a confirmed pathology and a note that still says "suspected", the treating physician is queried.

06What the documentation supports

• Rectal metastasis confirmed by the physician: C78.5, alongside the pancreatic primary (C25.-), which is still a documented, reportable condition.
• Second primary cancer of the rectum confirmed by the physician: C20, alongside the pancreatic cancer.
• Origin still uncertain: neither C78.5 nor C20. Code what is established, and the rectal mass to the highest degree of certainty for that encounter: in the Index, a mass of a specified organ leads to Disease, rectum, specified NEC (K62.89). D49.0 is not used: per the Tabular List, "mass", unless otherwise stated, is not regarded as a neoplastic growth.

07A suspicion is not a code

In a RADV validation, it is not enough for the diagnosis to appear on a problem list. The documentation has to identify a reportable condition, clinically supported and backed by an eligible encounter.

A suspicion can justify a diagnostic workup; it does not necessarily justify an HCC code.