In March 2017, CAP/AMP sent joint comment letters to MolDX and posted them publicly. In general, they feel that MolDX sets arcane and inappropriately high technical requirements for coverage tests. They assert that very, very few US labs could meet the closely fitted technical requirements found in the LCDs. They are concerned that NY State approval is required even for labs and patients in Midwestern or West Coast states. More after the break.
While "comprehensive genomic profiling" does not seem to be a copyrighted term, it is most often found in association with Foundation Medicine testing (e.g. here).
CAP/AMP also complain that MolDX requires CGP testing to be submitted with unlisted codes, whereas they argue that 81445/81455 CPT coding (gene panel tests, any genomic sequencing method or other genomic analysis method) are appropriate. AMA defines GSPs only as molecular analysis methods that "simultaneous assess multiple genes" and could intentionally include NGS, or PCR, or microarrays, etc. (see the publicly available AMA CPT preamble to GSP codes). CAP has also made this point at February 2017 and June 2017 AMA CPT public meetings. CAP/AMP notes that the NGS MAC uses the AMA CPT codes for somatic tumor gene panels, rather than an unlisted code.
By my math, Medicare use of Molecular Unlisted Code has been rising by more than doubling annually, and reached about 1/3 of all molecular services by 2015 (here).
CAP/AMP also note there is increasing evidence for the clinical value of doing paired germline/tumor sequencing. The MolDX LCDs remark that germline testing is excluded.
The AMP's Public Letters home page is here. Specific letters on melanoma here, on ovarian here, on colorectal cancer here. The three letters are 13, 7, and 8 pages long, respectively.