|
 |
 |
The first figure depicts a typical US population. Each point stands for the
number of people alive at a certain age. From the age of 30y and onward people
start dying and curve declines. From this curve it is possible to estimate
the probability to die at each age, which is called hazard rate and depicted
in the right figure. As we grow older our hazard rate (chance to die) continually
rises. I computed similar statistics for breast cancer and discovered that
the hazard rate of cancer behaves in a different way. It declines.
The data source
Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov)
SEER*Stat Database: Incidence - SEER 9 Regs Public-Use, Nov 2004 Sub (1973-2002),
National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics
Branch, released April 2005, based on the November 2004 submission.
The next figure depicts the relative survival of 313303 white females with
breast cancer along 30 years. The ordinate stands for the probability to live.
At the time of diagnosis the probability is one or 100%.
 |
 |
Cancer is a chronic disease
The declining curve illustrates that from the time of diagnosis and onward
women continue dying from cancer. However we ought to turn our attention to
the 60% which live with their cancer for about 25 years. Consider
a hypothetical woman who died 25 years after the diagnosis. Since dying from
cancer she obviously carried the disease for 25 years. What was her secret?
The survival curve tells us yet another interesting story. From year to year
its yearly decline becomes smaller. Compare it with a linearly declining
survival curve on the right whose rate of decline over the years rises. In
order to appreciate its significance we have to compute the hazard rate involved
in the survival curvs.
Hazard rate = (number of women dying during a particular year) / (number
of women alive at the beginning of this year)
The figure depicts the hazard rate with time of 313303 white women with breast
cancer
 |
At diagnosis the hazard rate is about 0.004 (A). Then it
rises and in the third year it reaches its peak of about 0.0055 (B). Then
it declines to a mean of about 0.045 (C) whereupon it rises again. When
diagnosed (A) breast cancer hazard rate is higher than that of the healthy
population. This bimodal curve clearly differs from that of the population
above which continually rises. Its steep ascent is associated with the
treatment which the patients received. From the third year and onward
(B) the patients lived with cancer better and better. However since the
disease progresses, it finally overpowers the patient and the hazard rate
rises again.
Cancer proceeds through four stages:
1. In situ: Regarded also as pre-cancer. 59578 patients
2. Localized: Tumor is confined to the breast. 249447 patients
3. Regional: Tumor has invaded local lymph nodes 134295 patients
4. Distant: Overt metastasis. 24870 patients
 |
The hazard rate of females with in situ cancer rises linearly in the
same way as in the healthy population. After all in situ cancer does not
kill. Localized cancer starts with a hazard rate = 0.0015 (A) which is
higher than that of in situ cancer (0.0005). In the coming years it rises
steeply and from the third year and onward (B) it rises slower than the
hazard of in situ cancer. In regional cancer the bi modality is most pronounced.
Since distant cancer hazard rates are relatively high it is depicted again.
 |
When distant cancer is first diagnosed it is decompensated and its hazard
rate = 0.035. Its decline is attributed to treatment. After reaching a minimum
at C it rises, mainly because the disease became resistant to treatment (D).
Localized cancer
A more refined analysis reveals the bi-modal pattern even in localized breast
cancer. The hazard of in situ cancer rises linearly. In localized cancer the
hazard starts ascending steeply and after the third year the slope declines
(left curve). In the right curve, in situ hazard was subtracted from the localized
and the bimodal pattern becomes somewhat clearer.
 |
.
Hypothesis
The patient depends somehow on her tumor and its removal raises the
hazard rate (A – B) which is most pronounced in compensated cancers
e.g., localized and regional. Initially the localized tumor is small and
by itself does not harm vital functions yet following its removal the
hazard rate rises steeply. From the third year (B) and onward the slope of
its hazard rate is slightly higher than that that of in situ cancers which
obviously do not harm vital functions. Many cancers diagnosed as regional
actually carry micrometastases which protect the patient from therapy
induced total tumor ablation. These patients survive after the third
year (B) and their hazard rate declines.
Cancer is caused by a deficiency of a yet unknown metabolite
A. In order to replenish the missing metabolite the organism grows a tumor
which produces a substitute B. Since the deficiency continually
aggravates, the tumor has to grow more and more in order to replenish
the missing metabolite. In advanced deficiency it destroys vital functions
and finally kills the patient. Tumor ablation aggravates the deficiency
and the hazard rises which is most pronounced in regional cancers. Patients
with micro metastases are protected from therapy induced total ablation and
their hazard rate declines.
Clinically this deficiency is manifested by a wasting disease which gradually
turns into overt cachexia and was named as pernicious
cachexia. The tumor is regarded here as a protective measure
against cachexia.
Treatment objectives: Do not treat unless the tumor causes pain and
distress or destroys vital functions. Wait as long as cancer is compensated
and treat only during decompensation.
Further reading
Pernicious cachexia
Hazard rates of other cancers
A model of this metabolic deficiency
Farewell my breast