First we present a chart on the increase in survival based on frequency of testing for ovarian cancer. This report is entitled "Genomic Tests for Ovarian Cancer Detection and Management" and was prepared for the Agency for Healthcare Research and Quality of HHS. It shows that an 80% reduction in mortality can be achieved if one screens every three months for ovarian cancer. Thus it is known what could be done. We will work through this approach again later.
Let us look at another view of this same problem. Let us start with a recent article in the journal Obstetrics & Gynecology: April 2009 - Volume 113 - Issue 4 - pp 775-782. The article is entitled: Results from four rounds of ovarian cancer screening in a randomized trial. The abstract states:
"OBJECTIVE: To test whether annual screening with transvaginal ultrasonography and CA 125 reduces ovarian cancer mortality.
METHODS: Data from the first four annual screens, denoted T0-T3, are reported. A CA 125 value at or above 35 units/mL or an abnormality on transvaginal ultrasonography was considered a positive screen. Diagnostic follow-up of positive screens was performed at the discretion of participants' physicians. Diagnostic procedures and cancers were tracked and verified through medical records.
RESULTS: Among 34,261 screening arm women without prior oophorectomy, compliance with screening ranged from 83.1% (T0) to 77.6% (T3). Screen positivity rates declined slightly with transvaginal ultrasonography, from 4.6 at T0 to 2.9-3.4 at T1-T3; CA 125 positivity rates (range 1.4-1.8%) showed no time trend. Eighty-nine invasive ovarian or peritoneal cancers were diagnosed; 60 were screen detected. The positive predictive value (PPV) and cancer yield per 10,000 women screened on the combination of tests were similar across screening rounds (range 1.0-1.3% for PPV and 4.7-6.2 for yield); however, the biopsy (surgery) rate among screen positives decreased from 34% at T0 to 15-20% at T1-T3. The overall ratio of surgeries to screen-detected cancers was 19.5:1. Seventy-two percent of screen-detected cases were late stage (III/IV).
CONCLUSION: Through four screening rounds, the ratio of surgeries to screen-detected cancers was high, and most cases were late stage. However, the effect of screening on mortality is as yet unknown."
This is a bit obtuse for the non-professional but it displays the standard approach to the study of many disease and the efficacy of procedures used to screen for their presence and the results of actions taken thereto. The question that the researchers went out to answer was the one which says did yearly screening for ovarian cancer have any benefit. We believe in a Fleckian manner that this question and the answer could be generalized by politicians and their ilk into one which is, is screening for ovarian cancer effective. They are two different questions. We have already shown above that they are effective.
Now let us look at the data from a different perspective. Namely, in contrast to the above study let us look at the underlying "physics" of the process and look at the facts and data as say an engineer would do. Here we go with the logic:
1. We know that the incidence of ovarian cancer is 14.4% in women 45-54, 21.4% in women 55-64, 25.3% in women 65-74, and 16.3% in women 75-84. (See Berek, Gynecology, 2008). Thus there are many women who will come down with this disease, a deadly disease if caught late.
2. The five year survival for ovarian cancer is 86% at State I, 70% at stage II, 34% at stage III and 19% at stage IV. (Schorge et al Gynecology 2008 p732). Thus if one can detect the cancer at State I it is possible under current means to have 86% or better survival. Stage I means growth limited to one or both ovaries with possible growth on the surface.
3. The ovary limitation means a tumor size of 2 to 4 cm diameter at most. That is the size of an ovary and it is also the size at which one can detect the lesion on ultrasound with some specificity. Using the CA125 at a level below 35 one may get better detection but higher false alarm rates. The problem with higher false alarm rates is that it requires surgery, and although it may be performed laporascopically at first, it may or may not require full laporotomy. The latter is the case if a malignancy is detected at surgery.
4. Cancer is a disease that starts with one aberrant cell. The call multiplies and attempts to double, each division, although that is not the case in reality for a variety of known and yet to be known reasons. However, 20 doublings can occur in less than one day, that is a total of 106 cells, not detectable. In 50 days we get to 40 doublings, or 1012 cells. By 125 days we get to 70 doublings, a bulky mass. (Weinberg, Cancer, p 365, 2008). However for many reasons due to the individual's immune system the doubling may take longer because there may be multiple genetic steps involved.
5. Cancer masses can be detected at 108 cells by imaging and at 109 cells by palpation. At 1012 cells the patient is on the road to death from the disease. (Weinberg, Cancer, p 363, 2008).
6. Thus if one performed the tests as described in the article every 120 days, then one may have a substantially improved chance of detecting the cancer at Stage I and achieving an 86% cure rate.
7. The current death rate from ovarian cancer is 8.6 per thousand females. This is a total of 280,000 women per year based upon CDC data.
8. If screening at 120 day intervals can reduce this to 42,000 deaths or equivalently save 238,000 women, at a cost of say $250 per screening or $1,000 per woman per year, over 45 years of age. The census states we have then we have a cost of 64.5 million women over 45. Thus it will cost $64.5 billion. Or, the cost per woman saved would be $271,000 per life saved per year.
First note that our simple analysis yields the same result as the HHS study we started with.
The question is it this worth it? What is a woman's life worth? Do we stop at say 65 or 75, do we continue to 85.
The other issue is that the authors of the article assumed annual testing. Based upon the logic above we see it means at least quarterly testing due to the tumor growth rate. By the way this applies to all tumors. Perhaps a study should address the question; "How frequently should testing be performed to obtain a material reduction in mortality from that disease". Clearly annual testing will at best get say one sixth of the cases; say 18% if everyone is tested.
This analysis has raised two issues:
1. When considering revising health care, what screening should be done and at what cost. Can, for example, a patient, person, pay for their own screening costs, at a price pari passu to the lowest cost paid, if they feel that they want more testing. Or will the Government as do the insurance companies today, have the lowest price forcing individual payers to subsidize the group payers, and in this case the Government. If the Government agrees to do annual testing and to be reliable it demands at least quarterly testing, then can a patient have the right to play on a level playing field or will the individual be taxed to seek better care on top of the costs?
2. When medical research is performed, there is a strong Fleck influence of a "thought collective" approach. The Fleck view of facts plays a significant role as well. The questions that should be posed are, "What level of screening result in what level of reduction in mortality?" Instead the way these are done is to take say an annual screening and determine if it is useful. The problem with this Fleckian "thought collective" approach is that it will then become part of the comparative clinical effectiveness schema as proposed by the Administration. Namely, the clinical result says that the screening is not useful. Wrong! The experiment shows that that specific type of screening is not useful.
Thus there are the above two issues of a much broader scope which can be drawn from this article, obscure as it may initially seem.
Posts
