How wrong was Hardin Jones?
While Hardin Jones was part right, his data and his overly pessimistic views are very much a product of his times.
One obvious indication of this is that the overall five year survival from breast cancer in the studies he examined was a dreadful 25 per cent, whereas NCI statistics for 2002 give an overall (all comers) five year survival of about 80% [4]. Some of this difference will be simply due to "lead time bias" i.e. longer survival simply due to most cases being diagnosed earlier in the course of their illness, but it illustrates the very different times.
Hardin Jones was working with data from very early in the last century, mainly from four studies containing the remarkable numbers of 651, 100, 64, and 100 untreated breast cancer patients along with a treated group. These were published in 1926 (!), 1937, (no reference given), and 1937 respectively. With the advanced cancers generally treated in those days, it is likely that most of these patients would be incurable even today.
None of the studies he examined were properly controlled trials. Well-planned clinical studies of any kind were still rare in those times, and it would also have been quite unethical even then to randomise very destructive cancers to a "no treatment" group, as demanded by any reliable comparison of treated with untreated patients.
Where, then, did the untreated patients come from? They were those who "refused operation or who had already advanced to an inoperable state". There are innumerable possible misleading influences in such studies. One is that breast cancers reaching a very advanced and inoperable state locally will include many slow growing, locally malignant cancers with little metastatic potential, where survival can be surprisingly long, if also made utterly miserable by enormous malignant masses, ulcers or cancer-en-cuirasse.
Nevertheless, as pointed out above, it is likely that even in the treated breast cancer groups the natural biology of the cancer was mainly determining length of survival (as applies with many cancers today), and treatment played a largely palliative role. It is also fairly certain in hindsight that the apparent survival benefits from treatment in some of the studies e.g. of cancer of the cervix were real effects, and Hardin Jones very reluctantly does allow this possibility.
The primary objective of medical treatment has always been to cure the patient permanently of their cancer, with palliation a close second. Hardin Jones was determined to argue that cancer always went its own way regardless of treatment, and dismissed contrary evidence whenever found. He even dismissed data favouring the cure of some subgroups of cancer patients, such as a survival curve identical to that of a normal population, by suggesting that they represented "cases with a milder type of disease than is usually reported" (p323). Even if true, that may simply mean that those cancers have been caught before they have been able to metastasise. That remains a primary objective in the treatment of solid cancers and the implications for the patient are the same.
Hardin Jones made a number of such judgements while never himself ever being involved in cancer treatment or clinical research. He was a physiologist and statistician, attached at the time to the Atomic Energy Commission in Berkeley, California.
Hardin Jones and Cancer