References are pretty easy to find:
After promising animal research, physician researchers connected with Yale University and the University of Chicago tried injecting sulfur mustard and its somewhat milder relatives, the nitrogen mustards, into the veins of cancer patients in the early 1940s. This was the first time a systemic treatment was given for cancer. “It would upset the whole body,” says Smith. “That proved to be valuable when you deal with certain kinds of leukemia or lymphoma.”
https://www.the-scientist.com/foundations/from-chemical-weap...
Not long after the discovery of nitrogen mustard, Sidney Farber of Boston demonstrated that aminopterin, a compound related to the vitamin folic acid, produced remissions in children with acute leukemia.
https://www.cancer.org/cancer/cancer-basics/history-of-cance...
I'm not certain Japan's Unit 731 was engaged in research with chemotherapy precursors, though the range of testing was extensive, and the US provided a grant of immunity after the war.
Please note that pretty much anything horrific you can imagine, and then some, is discussed in the links which follow.
https://nationalinterest.org/blog/reboot/unit-731-japans-sec...
I haven't followed up on this, but I seem to remember hearing about research about 20 years ago that even with cancer-associated mutations, cancer usually required physical damage to the intercellular matrix before it would manifest.
If that's true, I’m not sure identifying the genetic mutations as the cause (at least the proximate cause) of cancer might not be entirely accurate, or at least might be somewhat misleading.
Some years ago, a close friend died of a relatively rare and aggressive form of cancer. Among my roles was running interference against those who were pitching various miracle cures, with that tired old scam laetril playing a significant role.
In the decades since, I've watched developments in the science and development, as well as traced the history. It turns out that the treatment received a few decades ago was little changed from that of the 1960s, and has progressed little since. Genetic sequencing identified the affected chromosome of the triggering mutation in the 1980s, and we now know the specific codons involved. There remains no cure.
Five year survival remains about 20%. Laetril is still a fraud. And several of the other patients we came to know during the course of my friend's treatment have died since of conditions related to the treatment itself, though yes, they did buy another 10--20 years.
Cancer is brutal.
I will note: "cancer" is a symptom cluster manifested by malignant growth, and virtually always due to genetic mutation within the affected cells (I'd say "always" though I'm not completely sure of that). The specific triggers, and the possibilities for cure, vary tremendously. Some bladder cancers are virtually entirely treatable by, of all things, tuberculosis baccili. Skin cancers if caught early pose little risk. Others are tremendously aggressive, and an online friend's posts progressed from announcing their partner's diagnosis with liver cancer to the partner's death over the course of less than two months.
Broad-brush statements are virtually always false.