CMS has an NCD, 90.2, which was designed in 2017 (with help from senior policy folk at HHS) to provide an incentive for FDA approval of then-new NGS testing.
- FDA-approved NGS CDx tests were covered from the day of FDA approval, per the NCD.
- And any updates (new gene, new drug) were ALSO covered from the of approval.
But CMS has a hodge-podge of barriers. The 14-day rule delays inpatient surgical cases from getting comprehensive genomic profiling results (CGP) for weeks. If the oncologist has to order it, when he sees the patient well after the surgery event, that adds additional weeks. Too many patients get urgent but nonspecific chemo during the undue lags. Even worse, MACs may not start new coverage on "the day of FDA approval" until a year or more later - another "OMG" moment. (See here).
How could we make this mess worse?
At the December HCPCS meeting, CMS proposed new modifiers X1 and X2, to "implement" the NCD. I only see new problems with a solution that solves no existing problems.
Details follow.
##
Here's the PDF of the December HCPCS meeting:
https://www.cms.gov/files/document/b2-2025-public-meeting-agenda-december-17-18-2025.pdf
Here's page 96, agenda 41, the item that impacts NCD 90.2:
 |
| click to enlarge |
Decision Expected January 2026
But - Decision Not Out Yet
The January 2026 promised decision isn't out yet. Looking at dates, it may be that non-drug decisions come out slower (several months) than drug decisions (several weeks).
For me, as of 2/12/2026, I see the December 2025 final drug hcpcs narrativedecisions, but nothing yet in a separate column for December 2025 non-drug decisions.
Curiousity: X1 X2 In Use?
On the complete tables of all existing codes, I did notice that X1 and X2 seemed to be already in use for something called "broad" and "narrow" services. Here.
PROBLEMS WITH X1-X2 as PROPOSED
- Nothing Good: This accomplishes nothing, and is not known to be a problem.
- Staging Glitches: The X1-X2 language for staging doesn't apply to whole ranges of cancers (leukemia, lymphom, brain cancers, to name just 3 examples.)
- These apply to the NCD language and the X1-X2 language; groups have complained to CMS for a decade.
- Adjuvant Stage 2 Well Known: MolDx covers NGS for adjuvant therapy in select stage 2 cases of colon and breast cancer to name just 2 examples). Because if a cancer needs chemotherapy [adjuvant], it is "advanced.")
- Fit Both Codes: Some cancers will fit both codes (recurrent skin cancer may be stage 1 or 2, not stage 3 or 4. X1=Stage 2, X2=Recurrent).
- RNA SEQ: The NCD excludes NGS for RNA SEQ, which is increasingly common. The modifiers make no such allowance.
- ddPCR: MRD cases using NGS fall under these modifiers, but digital PCR MRD, which is increasingly common, don't.
- Like Useless KX Modifier: This accomplishes nothing. It's like the KX modifier, often used in DME, which says nothing but affirms the claim complies with rules. (General instructions ALREADY require that submitted service comply with rules.) A fraudulent provider would use -KX as easily as a legit one.
- X1 and GZ: Most any claim for stage 1 or 2 is nonpayable under the NCd, so it should already have a GZ modifier (this service or claim not reasonable and necessary.) Does CMS need both GZ and X1? Does X1 mean GZ isn't needed? Where does it say that/
- Side note: it looked to me like codes X1, X2, were already in use for something else.
