Your intuition and comparison to flu is correct. The genome of flu is miniscule and yet it mutates enough every year to render prior vaccines and infections less effective. "Cancer" is hundreds or thousands of different diseases with different mutations etc. It will be a marvel when market approval comes for a vaccine that eliminates a single type of cancer (e.g. a melanoma "vaccine"). A universal cancer vaccine will likely never happen. It's like having a single vaccine for every virus and bacteria on the planet.

Internal antigens are certainly investigated but yes, they are harder because they will only be exposed to the immune system in small pieces on MHC class 1 molecules and thus it can be even harder to differentiate tumor from normal. The checkpoint inhibitors enhance the immune system's ability to do this to an extent but they are non-specific boosters of immune function and can result in life-threatening autoimmune disease.

You said a programmable virus that you can program to target a specific tissue that was discovered in chicks. NDV is a virus in chickens that can be programmed to do that.

Yes, NDV (and other viral vector-based) cancer therapies are in development. They are not ready for primetime because it's a delicate balance between a viral vector that won't be immediately cleared by the immune system but also won't cause a severe infection and then you need a mechanism of action against the tumor that the virus can encode that isn't toxic to the person while also being effective at killing cancer cells. Also, unlike drugs, these viruses can potentially mutate which is another safety concern.

How about cancer transmission from a tapewrom to a human to give you some nightmares?

https://www.nejm.org/doi/10.1056/NEJMoa1505892

Newcastle Disease Virus?

Depends on the exact procedure. Blastomere biopsies are before that distinction.

EDIT: Had to edit my post because both blastomere and blastocyst biopsies happen on day 3 (blastomere in the early part of day 3, blastocyst later in day 3) but my overall point of "it depends on the exact nature of the test" is unchanged.

>Like if you were going to have 100 cells at that point and are now down to 90 or whatever

They do it even earlier than that, at like the 8 cell stage when presumably it is so early in embryogenesis that there's no impact. You are right that it is still a relatively new technology but so far there has yet to be evidence that it has any impact on the fetus beyond the IVF process itself.

In addition to NK-LGL like the other poster mentioned there is also Nasal Type Extranodal NK Cell lymphoma

They absolutely can and in fact there are a whole slew of them:

Angioimmuoblastic T-cell lymphoma

Anaplastic large cell lymphoma (ALK+ or ALK- as well as the breast cancer associated and primary cutaneous types)

Adult T cell lymphoma/leukemia

T-Acute Lymphoblastic Lymphoma/Leukemia

Nasal type extranodal T cell lymphoma

Hepatosplenic T-cell lymphoma

Enteropathy associated T-cell lymphoma

Monomorphic epitheliotropic intestinal T-cell lymphoma

T-Cell Prolymphocytic leukemia

Primary Cutaneous T-Cell Lymphomas (of which there are several different types like Mycosis Fungoides, Sezary Syndrome, Lymphomatoid papulosis, Primary cutaneous gamma/delta)

They are not as common as things like lung, colon, breast, or prostate cancer, even if you lump all of them together.

Source: pathologist (but not a hematopathologist so please don't ask me for more details, here is a free online "textbook" if you want to read more: https://www.pathologyoutlines.com/lymphoma.html)