In 1972, Luria was asked to set up and direct a new center for cancer research at MIT. The center would serve as an extension of MIT's flourishing molecular and cellular biology programs, and would be funded in part by the new National Cancer Program established at NIH. For the center, Luria chose an older factory building on the MIT campus, which was fairly close to the Biology Department, and leased at the time to a chocolate manufacturer. Luria spent about a year designing and supervising the remodeling of the former chocolate factory, becoming an expert, he said, in "the two major functions of a laboratory director: a knowledge of plumbing and the art of coaxing physical plant people to do in a week what usually might take two months." The Center for Cancer Research opened in the fall of 1973.
For his core staff, Luria chose two MIT colleagues--Phillips Robbins and David Baltimore--and recruited immunologist Herman Eisen, then chair of microbiology at Washington University School of Medicine. Under Luria's direction, researchers at the MIT Center for Cancer Research went on to make important discoveries. David Baltimore, looking at how cancer-causing viruses infect and permanently change healthy cells, identified the enzyme reverse transcriptase, which provided strong evidence for a process of RNA to DNA conversion within cells. He shared the 1975 Nobel Prize in Physiology or Medicine for this work. By the early 1980s, Robert Weinberg had isolated several oncogenes, the genes in cancer cells that produce the cancerous state. Later, Phillip Sharp, another early recruit, discovered that in some cells, messenger RNA, which mediates between DNA chains and the proteins they code for, is spliced so that some portions are removed and the "cut" ends reattached. Thus, the sequence of units in proteins is different from the sequence in the corresponding genes. For this discovery Sharp received the 1993 Nobel Prize in Physiology or Medicine. Reflecting on Luria's leadership of the center, Sharp later said, "Salva was a visionary who protected his young faculty from unnecessary interruptions, thus allowing their research programs to flourish in an ideal scientific environment. He was also a role model for how a scientist could shape and lead a community." Luria directed the center until his retirement in 1985.
In his later years at MIT, Luria, like many senior scientists, did progressively less hands-on research and more mentoring and supervising of graduate students. He continued to teach undergraduate biology courses, which he had always preferred to graduate seminars. (In 1975 a series of his basic biology lectures was published as 36 Lectures in Biology.) One graduate course he did enjoy teaching was a literary seminar for new biology graduate students. He had developed this to encourage them to keep reading widely even after entering the more specialized world of professional science, and also to explore the relevance of biological concepts to humanistic studies. Luria himself never lost sight of the intellectual worlds outside of science, and many of his articles and addresses during this period were reflections on the relationship between what C. P. Snow called "the two cultures" of science and the humanities.
In 1973 Luria published a biology book for popular audiences, titled Life--the Unfinished Experiment. The book, which received a National Book Award the following year, provided a simple view of biology in terms of molecular mechanisms, discussing evolution, heredity, genetics, and basic cell function. In the early 1980s the Alfred P. Sloan Foundation invited Luria to write an autobiography, to be included in a series of scientists' accounts of their lives in science. This was published as A Slot Machine, A Broken Test Tube in 1984.
Just before his retirement from MIT, Luria was appointed a senior scientist at Repligen, a biotechnology company in Cambridge, Massachusetts. He and his research associates Joan Suit and Jennifer Jackson were awarded a patent in 1990 for a developing a mutant strain of E. coli genetically engineered to produce a specific therapeutic protein more efficiently.
Luria died in Lexington, Massachusetts, on February 6, 1991, of a heart attack.