Abstract
The lessons are: (a) human cancers certainly respond to immunological manipulations. Efforts at human cancer immunotherapy are therefore worthwhile. (b) Prophylaxis is very different from therapy of pre-existing disease, and hence much enthusiasm should not be derived from successful prophylaxis studies. Even in case of infectious agents against which robust prophylaxis is routinely achieved, therapy is nearly impossible once the disease has established. (c) Studies with appropriate cancer models of mice and rats are useful. The notion that it is easy to cure cancers in mice is generally advanced the most confidently by those who have never cured a mouse of cancer by immunotherapy. (d) With a nod to James Carville, it is the antigen(s), stupid! We still do not know the identity of protective tumor antigens. If any lesson can be drawn at all, it may well be that cancer immunotherapy must move away from the one-shoe-fits-all therapeutic models of chemotherapy and must move to individualized approaches. (e) All targets are equal, but some are more equal than others. The key is specificity for cancer. That does not necessarily mean specificity for cancer cells. (f) Vaccitherapy must be attempted preferably in the minimal residual disease setting, even though this is certain to be time-taking and expensive. In the setting of bulky disease, vaccitherapy must be combined with blockade of inhibitory signals, or depletion of down-regulatory T cells. Inhibition of effector level suppression of immune response is a key. Vaccitherapy alone or immuno-modulation alone is unlikely to succeed in therapy of bulky metastatic disease.