I apologize for the size of this post. But I think that many will agree that the serious nature of this (non-Tesla-specific) topic in particular deserves adequate explanation and documentation.
Background
In general, we are involved with this online club because we like cars; Tesla vehicles in particular. But it is probably a truism to say that you can't really enjoy your car/hobby if your health is not good.
Sadly, many of us over a certain age have experienced the death of family members, friends, or work acquaintances due to cancer. For many forms of cancer (~200 types) early detection is still difficult. Yet early detection often gives patients the best chance for success in battling dangerous cancer varieties (World Health Organization). “Just four types of cancer—breast, cervical, colorectal, and lung—currently have a screening test recommended by the United States Preventive Services Task Force (USPSTF). These cancers make up [only] 29% of all U.S. cancer cases.” (NORC at University of Chicago)
Fortunately, with the advent of new technologies this may improve dramatically in the near future (say within as little as five years). “Multi-cancer early detection (MCED) tests have captured the attention of doctors, researchers and patients due to their potential to detect a range of cancers through a simple blood test…MCED tests, a type of liquid biopsy, aim to [detect] early-stage cancer cells [in the blood stream] long before symptoms appear.” (MD Anderson Cancer Center at U-Texas) There seem to be a number of efforts ongoing to develop multi-cancer early detection tests (for example, see here).
Reason for this Post
This post--by me, a TMC member and a non-paid, not-a-medical-professional--is intended to inform readers of one such promising test: (the "Galleri" MCED test by the medical firm Grail. (See list of cancer types that may potentially be detected in Appendix A). Its purpose is also to inform interested readers how they may be able to receive the test free-of-charge as part of the ongoing Pathfinder 2 study.
According to the Grail website: “The Galleri test is recommended for use in adults with an elevated risk for cancer, such as those aged 50 or older. The Galleri test does not detect all cancers and should be used in addition to routine cancer screening tests recommended by a healthcare provider. Galleri is intended to detect cancer signals and predict where in the body the cancer signal is located. Use of Galleri is not recommended in individuals who are pregnant, 21 years old or younger, or undergoing active cancer treatment.
While the Galleri test is currently available (for ~$960), its widespread use has not yet been fully approved by appropriate (USA) government regulatory agencies. Of course it is not yet covered by any insurance. However, an option for taking the test (which I myself did) is to participate in the active Pathfinder 2 clinical study, which is still accepting qualified participants, (e.g., those belonging to valid U.S. healthcare systems, in certain areas, over the age of 49, not currently being evaluated/treated for cancer, and willing to answer easy questionnaires over a two to three year period). The incentive for those tested? The Galleri blood test itself is free for study participants. Also, as I understand it, Pathfinder 2 study results are not shared with other health care providers (so chances of later insurance rejection due to a "pre-existing condition" are probably low).
There is no “placebo” as I understand it; all applicable participants receive the actual Galleri test of their blood samples. Instead of a traditional biopsy (wherein tissue from a suspected tumor is surgically removed and visually analyzed under a microscope by technicians trained to spot aberrant cells), Galleri is a so-called “liquid biopsy"--wherein a small, easily-acquired blood sample is analyzed for genetic/biochemical markers that may indicate the possibility of certain types of cancer developing in various organs of the body. Results take about ten days, and are in the form of a “negative” (no cancer detected) or a “positive” (cancer detected, hopefully with information about the likely organ of origin). Probably because they comprise a non-random sample, the vast majority (98-99%) of Pathfinder 2 Study participants tested so far have received negative results. The relatively few test-takers with positive test results may receive (free-of-charge, I believe) follow-up diagnostic consultation and scans that attempt to help confirm (or refute) the actual presence and organ of origin of any suspected cancer.
What Does the Company, Grail, Get?
On the face of it, this sounds like a pretty good deal for study participants. So what does Grail receive in return? The answer: valuable help in establishing statistical validity (as well as feedback from participants on any emotional impacts; see Update below). Grail researchers naturally want to pin down the mathematical accuracy of the test. How reasonably sure can researchers, patients, and their physicians be of test results? Can levels of error be accurately quantified?
By providing its cutting-edge MCED test to an ever increasing number of volunteers and then following their subsequent histories, researchers should be able to more accurately quantify the level of statistical assurance of Galleri Test results. (See Appendix B for a discussion of statistical error.)
Potential Drawbacks
As for most things in life, there are potential drawbacks for study participants as well as advantages:
In my humble opinion, advantages of MCED tests test like Galleri far out-weigh potential drawbacks, even for a “positive” result. Besides gaining critical medical information sooner than was normally the case, there is also the opportunity to help (in a small way) refine what may eventually become a critically important medical tool in the fight against cancer. One might say that participation in this multi-cancer early-detection study is somewhat similar/analogous to involvement in Tesla’s Full Self Driving beta software program (see a list of proposed similarities in Appendix C). While both new technologies are still being perfected, they happen to be available in preliminary forms now, and with proper maturation may someday save a great many lives.
However, I can completely understand the natural tendency not to want to take a test, no matter how simple, that could potentially find a dangerous disease, especially when there are no other obvious symptoms. Why go looking for trouble? One answer is: to find trouble (and kick its butt) before it does the same to you.
Clearly every prospective participant should (along with their physician) carefully consider the meaning, use, and potential impacts of any medical test or procedure. Be aware that there can be genuine drawbacks from widespread, comprehensive cancer screening that may not be immediately obvious. For example, patients risk unnecessary, "over-treatment" for certain kinds of cancer. Apparently, for-profit medical organizations exist that promote cancer treatment that may not always be necessary. Not all cancer varieties are created equal and some thoughtful physicians advocate caution to avoid over-screening, unnecessary diagnosis, and over-treatment of certain "relatively harmless" forms of cancer. ("Relatively harmless" in that patients are much more likely to pass away from other causes before a slow-growing cancer can advance enough to cause death. For more on this, see Appendix D.) Nonetheless, early meaningful detection of truly dangerous and, once identified, potentially treatable forms of cancer remains a medical "holy grail." (Could that expression have prompted the origin of the "Galleri" Test company's name?)
OK, I'm Sold. So How Do I Participate?
If interested, please consult your primary health care physician or your greater health care organization--or search online for "Grail," "Galleri," or "Pathfinder 2"--to determine whether you qualify to participate in the Pathfinder 2 (or any other cancer-screening study) in your area.
Disclosure
I am over 65, living in the Sacramento California area, fortunately enjoying relatively good health, and benefiting from better-than-average health care coverage. I connected to a regional Pathfinder 2 Study group component through the Sutter Health medical group. (Luckily, I received a “negative” Galleri test result.) I have no out-of-the-ordinary business or financial relationships with any of the organizations or businesses discussed, except as a normal (occasional) patient. As a Pathfinder 2 study participant I was an unpaid volunteer, but like other study participants I received the Galleri blood sample test free of charge. (Since I was “negative,” free follow-up diagnostic scans were unnecessary in my case.) I have received, am receiving, and will receive no other compensation of any kind for posting this. As stated elsewhere, other companies and organizations, besides Grail, are working on similar technologies (for example, see here).
******************************************************************************************************************************************************************
******************************************************************************************************************************************************************
In Defense of "Science"
The concept of “science” has evolved over many hundreds of years or more (in various world regions and cultures) to become today a very effective and accurate--though often painstaking, time-consuming, and thereby at times frustrating--process to help determine the nature of matter, energy, behavior, ecology, and many other aspects of our lives; from the very large down to the very small, from the very fast to the very slow, and from the very near to the very far away.
Progressive historical experience has taught scientists to make observations, ask questions, review prior pertinent work, develop hypotheses (formal conjectures), perform experiments and studies, and formally report fully-documented findings, along with new as yet untested predictions, in academic and professional journals subject to critical review. In school students are often taught to envision scientific research as investigating whether the so-called "null" hypothesis (no effect; denoted H0) is true, or whether one or more of any number of "alternative" hypotheses (H1, H2, etc.; or otherwise Ha) are more valid for explaining observations.
In practice the best scientists are both enthusiastic and energetic, and yet as unbiased, objective, and down to earth as possible. Facts, evidence, and probabilities--along with imagination and hard work---remain important concepts in science. Engineering (e.g., the design, construction, and maintenance of manufactured products, materials, vehicles, buildings, power grids, conveyance structures, computers, electronics, and the myriad of other items our societies depend on) is one outcome of applied science. Medicine is another. Mathematics and Statistics are numerical tools used within the theoretical and applied sciences to--among other things--quantify, manipulate, model, evaluate, and validate data and conclusions. Inherent to science is an attempt to uncover the truth. What is the true nature of our world and of the greater universe?
Statistical Error
Studies and experiments are naturally subject to occasional error. Scientists, engineers, and physicians attempt to minimize the likelihood of error through careful experimental design, accurate data collection, and rigorous statistical analysis. One way they validate their conclusions is by reporting the likely level of statistical error in study reports. In many disciplines researchers traditionally hoped to achieve at least a 95 (or even 99) percent level of statistical assurance for their experimental findings. Reports routinely included calculated α (alpha) values—the likelihood of a Type I (false positive) error—along with appropriate statistical parameters (e.g., mean, standard deviation, regression, T-test, analysis of variance, and the many other parametric and nonparametric statistical calculations) in articles published in professional journals.
While in scientific research focus was frequently on reducing the likelihood of false positives, in the real world the likelihood of a Type II error, symbolized by β (beta) and meaning the chance of reporting a false negative result, may be of equal concern.
Unfortunately, the two error rates are apparently not directly linked mathematically in a satisfyingly simple way. Many attempts to reduce one type of error may be accomplished only at the expense of raising the likelihood of the second error type. One way to partly manage both α and β simultaneously has been by significantly increasing the number of subjects tested or measurements taken (i.e., the “sample size” in an experiment or study). But repeating experiments over and over, taking large numbers of measurements, and testing large numbers of subjects in real world scenarios can be labor-intensive, time consuming, and expensive. Nonetheless, good medical research is expected to go the extra distance for critical studies, especially where the health and well-being of the public is at stake.
Evaluating Error
On a simple level, results from the Galleri test may be considered “binomial.” That is, at first there appear to be only two possible outcomes: (a) evidence of cancer or (b) no such evidence. But there is more to it than that. When one considers the true (but initially unknown) nature of the patient’s actual condition (which the Galleri test is of course attempting to reveal) there are actually four potential outcomes:
The two types of error (False Positive or Type I, and False Negative or Type II) are indicated by red squares in the logical-square diagram above. One objective of the current Pathfinder 2 study is accurately pin down (quantify) the likelihoods of Type I and Type II errors associated with the Galleri Test. Clearly, it is advantageous for error rates to be both (a) identified and (b) minimized.
******************************************************************************************************************************************************************
******************************************************************************************************************************************************************
Another Viewpoint
Understandably, a diagnosis of cancer has traditionally invoked fear, panic, and other intense reactions. Despite important advancements, “In 2021, there will be an estimated 1.9 million new cancer cases diagnosed and 608,570 cancer deaths in the United States.” (source) Despite these sobering numbers, should cancer always be looked for and, if found, always removed from the body at any cost? Some experts are saying no; not just to save time and money, but because a less drastic approach (e.g., wait-and-see monitoring) may sometimes be better for patients.
One active advocate for a more responsible approach to cancer screening and treatment is Dr. H. Gilbert Welch. (See his video.) (Side note: As I recall, another advocate for this type of thinking was oncologist Dr. Geoffrey Sonn of Stanford.)
In his video Dr. Welch uses a simplistic "bird, rabbit, and turtle" analogy to describe different categories of "cancer." (According to Dr. Welch, many patients and even some medical professionals react to the words "cancer" and (separately) to "infection" the same way; as if the threat and outcome are always the same, when in truth they are not.) Per Welch, different kinds of cancer, in patients of different ages and health conditions, should require different levels of response. A more appropriate approach to oncology should include careful consideration of the nature of the disease; the patient-specific, non-cancer-related life-expectancy; and the likelihood of serious impacts (unintended consequences & side-effects) from aggressive cancer treatment. None of this sounds particularly new or earth-shaking; yet according to Welch some groups are still taking advantage of patient vulnerabilities to push treatments (sometimes at exotic locales) that may not, in fact, really be necessary.
Despite reasonable arguments in favor of a more moderate approach to screening for and treating cancer, some health facilities continue to promote aggressive (and expensive) treatment plans, intending to to rid patients of cancer at any cost. Is this always the wisest course?
Criticism
Also, please be aware that critics exist of the alleged "special treatment" the Galleri test is receiving in some quarters. For example, here.
In Conclusion
Although I applaud the development of multi-cancer early detection tests, I also recommend that interested readers review Dr. Welch's video before considering any cancer detection test--all the better to respond to any possible "positive" result.
*****
Background
In general, we are involved with this online club because we like cars; Tesla vehicles in particular. But it is probably a truism to say that you can't really enjoy your car/hobby if your health is not good.
Sadly, many of us over a certain age have experienced the death of family members, friends, or work acquaintances due to cancer. For many forms of cancer (~200 types) early detection is still difficult. Yet early detection often gives patients the best chance for success in battling dangerous cancer varieties (World Health Organization). “Just four types of cancer—breast, cervical, colorectal, and lung—currently have a screening test recommended by the United States Preventive Services Task Force (USPSTF). These cancers make up [only] 29% of all U.S. cancer cases.” (NORC at University of Chicago)
Fortunately, with the advent of new technologies this may improve dramatically in the near future (say within as little as five years). “Multi-cancer early detection (MCED) tests have captured the attention of doctors, researchers and patients due to their potential to detect a range of cancers through a simple blood test…MCED tests, a type of liquid biopsy, aim to [detect] early-stage cancer cells [in the blood stream] long before symptoms appear.” (MD Anderson Cancer Center at U-Texas) There seem to be a number of efforts ongoing to develop multi-cancer early detection tests (for example, see here).
Reason for this Post
This post--by me, a TMC member and a non-paid, not-a-medical-professional--is intended to inform readers of one such promising test: (the "Galleri" MCED test by the medical firm Grail. (See list of cancer types that may potentially be detected in Appendix A). Its purpose is also to inform interested readers how they may be able to receive the test free-of-charge as part of the ongoing Pathfinder 2 study.
According to the Grail website: “The Galleri test is recommended for use in adults with an elevated risk for cancer, such as those aged 50 or older. The Galleri test does not detect all cancers and should be used in addition to routine cancer screening tests recommended by a healthcare provider. Galleri is intended to detect cancer signals and predict where in the body the cancer signal is located. Use of Galleri is not recommended in individuals who are pregnant, 21 years old or younger, or undergoing active cancer treatment.
While the Galleri test is currently available (for ~$960), its widespread use has not yet been fully approved by appropriate (USA) government regulatory agencies. Of course it is not yet covered by any insurance. However, an option for taking the test (which I myself did) is to participate in the active Pathfinder 2 clinical study, which is still accepting qualified participants, (e.g., those belonging to valid U.S. healthcare systems, in certain areas, over the age of 49, not currently being evaluated/treated for cancer, and willing to answer easy questionnaires over a two to three year period). The incentive for those tested? The Galleri blood test itself is free for study participants. Also, as I understand it, Pathfinder 2 study results are not shared with other health care providers (so chances of later insurance rejection due to a "pre-existing condition" are probably low).
There is no “placebo” as I understand it; all applicable participants receive the actual Galleri test of their blood samples. Instead of a traditional biopsy (wherein tissue from a suspected tumor is surgically removed and visually analyzed under a microscope by technicians trained to spot aberrant cells), Galleri is a so-called “liquid biopsy"--wherein a small, easily-acquired blood sample is analyzed for genetic/biochemical markers that may indicate the possibility of certain types of cancer developing in various organs of the body. Results take about ten days, and are in the form of a “negative” (no cancer detected) or a “positive” (cancer detected, hopefully with information about the likely organ of origin). Probably because they comprise a non-random sample, the vast majority (98-99%) of Pathfinder 2 Study participants tested so far have received negative results. The relatively few test-takers with positive test results may receive (free-of-charge, I believe) follow-up diagnostic consultation and scans that attempt to help confirm (or refute) the actual presence and organ of origin of any suspected cancer.
What Does the Company, Grail, Get?
On the face of it, this sounds like a pretty good deal for study participants. So what does Grail receive in return? The answer: valuable help in establishing statistical validity (as well as feedback from participants on any emotional impacts; see Update below). Grail researchers naturally want to pin down the mathematical accuracy of the test. How reasonably sure can researchers, patients, and their physicians be of test results? Can levels of error be accurately quantified?
By providing its cutting-edge MCED test to an ever increasing number of volunteers and then following their subsequent histories, researchers should be able to more accurately quantify the level of statistical assurance of Galleri Test results. (See Appendix B for a discussion of statistical error.)
Update: Based on the first post-test questionnaire, Grail is also interested in the mental/emotional effects (if any) on participants of receiving an early-detection test for cancer. Feedback will probably help Grail's eventual marketing of the test product.
Potential Drawbacks
As for most things in life, there are potential drawbacks for study participants as well as advantages:
- Time required; in most cases relatively minimal. Participants must provide a simple blood sample (collected in just a few small vials) at a participating phlebotomist facility and answer (up to 4) questionnaires (via the Internet) over approximately two to three years.
- Heightened anxiety. Waiting for and pondering test results may cause worry in some participants. Are both negative and positive results to be believed? The evidence so far is that the Galleri MCED test is reasonably accurate; however as in most medical and other scientific tests “false positive” and “false negative” results may occur. The actual error rates are, I suspect, what the Pathfinder 2 study is in part intended to help determine.
|- False positive results (an indication of cancer where none exists) might be expected to initially cause (unwarranted) alarm, since the error cannot immediately be known. But timely follow-up diagnostic tests (normally provided free of charge) should eventually help allay concerns in most subjects with positive results who in truth do not have cancer.
- False negative results (failure to identify an existing cancer) are also cause for concern. In such a hypothetical situation patients initially unaware of an actual developing disease would unfortunately not be alerted otherwise via the Galleri test. Although continuing ignorance of an otherwise potentially dangerous condition was pre-existing, incautious participants might be inappropriately lulled into a false state of (oncological) security by a false negative result.
- False positive results (an indication of cancer where none exists) might be expected to initially cause (unwarranted) alarm, since the error cannot immediately be known. But timely follow-up diagnostic tests (normally provided free of charge) should eventually help allay concerns in most subjects with positive results who in truth do not have cancer.
To combat both these (hopefully rare) erroneous possibilities, study officials should take steps to educate participants (e.g., through their physicians) beforehand. To start with, potential study participants should be clear that the Galleri test is still being developed and evaluated. Probably few medical tests are 100% foolproof. Indeed, the very intent of the Pathfinder 2 study is to try to better pinpoint accuracy of test results. Participants should be well aware of and comfortable with these conditions and limitations.
And study participants, too, must take responsibility. Wise participants will not let either positive or negative results cloud their otherwise good lifestyle judgements. They should continue to eat a proper, healthy diet; refrain from smoking; get proper exercise; minimize or eliminate alcohol, red meat, and refined sugar; regularly consume plenty of fresh fruits and vegetables; and continue to take advantage of all appropriate recommended medical consultations and cancer-screening opportunities (e.g., colonoscopies, PSA blood tests, mammograms, PAP smears, and so forth).
- As mentioned, the Galleri Test currently identifies only about 50 types of cancer. (But included in that group are a number of very serious diseases. And as technology improves, the range of cancer types covered by testing may always widen for this or other MCED tests in the future.)
| - And, of course, in rare instances the unpleasant necessity for cancer treatment, with all that entails. Despite such a "drawback," early-detection by the Galleri test may hopefully help minimize treatment costs and efforts and--best of all--give patients the best chance for extended life and perhaps full recovery.
In my humble opinion, advantages of MCED tests test like Galleri far out-weigh potential drawbacks, even for a “positive” result. Besides gaining critical medical information sooner than was normally the case, there is also the opportunity to help (in a small way) refine what may eventually become a critically important medical tool in the fight against cancer. One might say that participation in this multi-cancer early-detection study is somewhat similar/analogous to involvement in Tesla’s Full Self Driving beta software program (see a list of proposed similarities in Appendix C). While both new technologies are still being perfected, they happen to be available in preliminary forms now, and with proper maturation may someday save a great many lives.
However, I can completely understand the natural tendency not to want to take a test, no matter how simple, that could potentially find a dangerous disease, especially when there are no other obvious symptoms. Why go looking for trouble? One answer is: to find trouble (and kick its butt) before it does the same to you.
Clearly every prospective participant should (along with their physician) carefully consider the meaning, use, and potential impacts of any medical test or procedure. Be aware that there can be genuine drawbacks from widespread, comprehensive cancer screening that may not be immediately obvious. For example, patients risk unnecessary, "over-treatment" for certain kinds of cancer. Apparently, for-profit medical organizations exist that promote cancer treatment that may not always be necessary. Not all cancer varieties are created equal and some thoughtful physicians advocate caution to avoid over-screening, unnecessary diagnosis, and over-treatment of certain "relatively harmless" forms of cancer. ("Relatively harmless" in that patients are much more likely to pass away from other causes before a slow-growing cancer can advance enough to cause death. For more on this, see Appendix D.) Nonetheless, early meaningful detection of truly dangerous and, once identified, potentially treatable forms of cancer remains a medical "holy grail." (Could that expression have prompted the origin of the "Galleri" Test company's name?)
OK, I'm Sold. So How Do I Participate?
If interested, please consult your primary health care physician or your greater health care organization--or search online for "Grail," "Galleri," or "Pathfinder 2"--to determine whether you qualify to participate in the Pathfinder 2 (or any other cancer-screening study) in your area.
Disclosure
I am over 65, living in the Sacramento California area, fortunately enjoying relatively good health, and benefiting from better-than-average health care coverage. I connected to a regional Pathfinder 2 Study group component through the Sutter Health medical group. (Luckily, I received a “negative” Galleri test result.) I have no out-of-the-ordinary business or financial relationships with any of the organizations or businesses discussed, except as a normal (occasional) patient. As a Pathfinder 2 study participant I was an unpaid volunteer, but like other study participants I received the Galleri blood sample test free of charge. (Since I was “negative,” free follow-up diagnostic scans were unnecessary in my case.) I have received, am receiving, and will receive no other compensation of any kind for posting this. As stated elsewhere, other companies and organizations, besides Grail, are working on similar technologies (for example, see here).
******************************************************************************************************************************************************************
- Adrenal Cortical Carcinoma
- Ampulla of Vater
- Anus
- Appendix, Carcinoma
- Bile Ducts, Distal
- Bile Ducts, Intrahepatic
- Bile Ducts, Perihilar
- Bladder, Urinary
- Bone
- Breast
- Cervix
- Colon and Rectum
- Esophagus and Esophagogastric Junction
- Gallbladder
- Gastrointestinal Stromal Tumor
- Gestational Trophoblastic Neoplasms
- Kidney
- Larynx
- Leukemia
- Liver
- Lung
- Lymphoma (Hodgkin and Non-Hodgkin)
- Melanoma of the Skin
- Merkel Cell Carcinoma
- Mesothelioma, Malignant Pleural
- Nasal Cavity and Paranasal Sinuses
- Nasopharynx
- Neuroendocrine Tumors of the Appendix
- Neuroendocrine Tumors of the Colon and Rectum
- Neuroendocrine Tumors of the Pancreas
- Oral Cavity
- Oropharynx (HPV-Mediated, p16+)
- Oropharynx (p16-) and Hypopharynx
- Ovary, Fallopian Tube and Primary Peritoneum
- Pancreas, exocrine
- Penis
- Plasma Cell Myeloma and Plasma Cell Disorders
- Prostate
- Small Intestine
- Soft Tissue Sarcoma of the Abdomen and Thoracic Visceral Organs
- Soft Tissue Sarcoma of the Head and Neck
- Soft Tissue Sarcoma of the Retroperitoneum
- Soft Tissue Sarcoma of the Trunk and Extremities
- Soft Tissue Sarcoma Unusual Histologies and Sites
- Stomach
- Testis
- Ureter, Renal Pelvis
- Uterus, Carcinoma and Carcinosarcoma
- Uterus, Sarcoma
- Vagina
- Vulva
******************************************************************************************************************************************************************
Appendix B
Statistical Error
Statistical Error
In Defense of "Science"
The concept of “science” has evolved over many hundreds of years or more (in various world regions and cultures) to become today a very effective and accurate--though often painstaking, time-consuming, and thereby at times frustrating--process to help determine the nature of matter, energy, behavior, ecology, and many other aspects of our lives; from the very large down to the very small, from the very fast to the very slow, and from the very near to the very far away.
Progressive historical experience has taught scientists to make observations, ask questions, review prior pertinent work, develop hypotheses (formal conjectures), perform experiments and studies, and formally report fully-documented findings, along with new as yet untested predictions, in academic and professional journals subject to critical review. In school students are often taught to envision scientific research as investigating whether the so-called "null" hypothesis (no effect; denoted H0) is true, or whether one or more of any number of "alternative" hypotheses (H1, H2, etc.; or otherwise Ha) are more valid for explaining observations.
In practice the best scientists are both enthusiastic and energetic, and yet as unbiased, objective, and down to earth as possible. Facts, evidence, and probabilities--along with imagination and hard work---remain important concepts in science. Engineering (e.g., the design, construction, and maintenance of manufactured products, materials, vehicles, buildings, power grids, conveyance structures, computers, electronics, and the myriad of other items our societies depend on) is one outcome of applied science. Medicine is another. Mathematics and Statistics are numerical tools used within the theoretical and applied sciences to--among other things--quantify, manipulate, model, evaluate, and validate data and conclusions. Inherent to science is an attempt to uncover the truth. What is the true nature of our world and of the greater universe?
Statistical Error
Studies and experiments are naturally subject to occasional error. Scientists, engineers, and physicians attempt to minimize the likelihood of error through careful experimental design, accurate data collection, and rigorous statistical analysis. One way they validate their conclusions is by reporting the likely level of statistical error in study reports. In many disciplines researchers traditionally hoped to achieve at least a 95 (or even 99) percent level of statistical assurance for their experimental findings. Reports routinely included calculated α (alpha) values—the likelihood of a Type I (false positive) error—along with appropriate statistical parameters (e.g., mean, standard deviation, regression, T-test, analysis of variance, and the many other parametric and nonparametric statistical calculations) in articles published in professional journals.
While in scientific research focus was frequently on reducing the likelihood of false positives, in the real world the likelihood of a Type II error, symbolized by β (beta) and meaning the chance of reporting a false negative result, may be of equal concern.
Unfortunately, the two error rates are apparently not directly linked mathematically in a satisfyingly simple way. Many attempts to reduce one type of error may be accomplished only at the expense of raising the likelihood of the second error type. One way to partly manage both α and β simultaneously has been by significantly increasing the number of subjects tested or measurements taken (i.e., the “sample size” in an experiment or study). But repeating experiments over and over, taking large numbers of measurements, and testing large numbers of subjects in real world scenarios can be labor-intensive, time consuming, and expensive. Nonetheless, good medical research is expected to go the extra distance for critical studies, especially where the health and well-being of the public is at stake.
Evaluating Error
On a simple level, results from the Galleri test may be considered “binomial.” That is, at first there appear to be only two possible outcomes: (a) evidence of cancer or (b) no such evidence. But there is more to it than that. When one considers the true (but initially unknown) nature of the patient’s actual condition (which the Galleri test is of course attempting to reveal) there are actually four potential outcomes:
- A true “positive” result. (The patient truly has cancer and the test reveals it.)
- A true “negative” result. (There is no cancer and test results reflect that.)
- A false “positive” result. (Test results indicate cancer where none exists.)
- A false “negative” result. (The test finds no cancer when it is actually present.)
|
- The immediate results of the Galleri test (positive or negative).
- And any subsequent initially-undetected presence of pertinent cancer in study participants.
The two types of error (False Positive or Type I, and False Negative or Type II) are indicated by red squares in the logical-square diagram above. One objective of the current Pathfinder 2 study is accurately pin down (quantify) the likelihoods of Type I and Type II errors associated with the Galleri Test. Clearly, it is advantageous for error rates to be both (a) identified and (b) minimized.
******************************************************************************************************************************************************************
Appendix C
Imagined Similarities Between the Tesla Full Self Driving (FSD) Beta Program
and the Grail Galleri MCED Test Pathfinder 2 Medical Study
Imagined Similarities Between the Tesla Full Self Driving (FSD) Beta Program
and the Grail Galleri MCED Test Pathfinder 2 Medical Study
- Both are programs involve new, ground-breaking, “high-tech” procedures.
- There is no “placebo.” Participants receive the actual product.
- In both cases there are competing firms, research organizations, and/or agencies in various stages of developing similar products intended for sale to the general public.
- Both companies—Tesla and Grail—are for-profit businesses that have invested large sums and much effort to design, develop, and bring to market safe and effective products.
- Both products are intended to improve and to save human life.
- Despite important progress made so far, products from both programs should probably be regarded as still "under development."
- Both companies have asked for qualified, responsible volunteers among the public to actively test their products “on the fly,” while development continues and before final governmental approval is issued.
- While drawbacks exist and further development is arguably needed, these promising yet still emerging technologies may nonetheless benefit users right now.
******************************************************************************************************************************************************************
Appendix D
Putting Cancer Screening, Diagnosis, and
Treatment into More Proper Perspective
Putting Cancer Screening, Diagnosis, and
Treatment into More Proper Perspective
Another Viewpoint
Understandably, a diagnosis of cancer has traditionally invoked fear, panic, and other intense reactions. Despite important advancements, “In 2021, there will be an estimated 1.9 million new cancer cases diagnosed and 608,570 cancer deaths in the United States.” (source) Despite these sobering numbers, should cancer always be looked for and, if found, always removed from the body at any cost? Some experts are saying no; not just to save time and money, but because a less drastic approach (e.g., wait-and-see monitoring) may sometimes be better for patients.
One active advocate for a more responsible approach to cancer screening and treatment is Dr. H. Gilbert Welch. (See his video.) (Side note: As I recall, another advocate for this type of thinking was oncologist Dr. Geoffrey Sonn of Stanford.)
In his video Dr. Welch uses a simplistic "bird, rabbit, and turtle" analogy to describe different categories of "cancer." (According to Dr. Welch, many patients and even some medical professionals react to the words "cancer" and (separately) to "infection" the same way; as if the threat and outcome are always the same, when in truth they are not.) Per Welch, different kinds of cancer, in patients of different ages and health conditions, should require different levels of response. A more appropriate approach to oncology should include careful consideration of the nature of the disease; the patient-specific, non-cancer-related life-expectancy; and the likelihood of serious impacts (unintended consequences & side-effects) from aggressive cancer treatment. None of this sounds particularly new or earth-shaking; yet according to Welch some groups are still taking advantage of patient vulnerabilities to push treatments (sometimes at exotic locales) that may not, in fact, really be necessary.
Despite reasonable arguments in favor of a more moderate approach to screening for and treating cancer, some health facilities continue to promote aggressive (and expensive) treatment plans, intending to to rid patients of cancer at any cost. Is this always the wisest course?
Criticism
Also, please be aware that critics exist of the alleged "special treatment" the Galleri test is receiving in some quarters. For example, here.
In Conclusion
Although I applaud the development of multi-cancer early detection tests, I also recommend that interested readers review Dr. Welch's video before considering any cancer detection test--all the better to respond to any possible "positive" result.
- Detection and treatment of "cancer" does not (and should not) always mean the same thing in all patients.
- There are many (~200) different types of cancers and they may differ greatly in speed of growth, aggressiveness, rates of mortality, treatment impacts, and the likelihood of ultimate treatment "success."
- And what does "success" actually mean for any particular patient? Is it total cancer eradication? (Naturally, that is the goal in many cases.) Is it continuation of life for some limited period? Avoidance of pain? Death with dignity? (These may be unpleasant, but are probably important, questions.)
- Plans for any regular cancer screening and treatment might want to take into account such factors as anticipated life-expectancy of the patient and likely impacts of treatment.
- Dr. Welch points out that certain fast-advancing, aggressive cancers may not be helped even by early screening, while certain slow-growing cancers do not tend to eventually cause mortality (in older patients).
- What I take away from the video is that--subject to patient age and general health--patients with moderate-growth-rate dangerous cancers may tend to benefit most from early screening (and subsequent treatment).
- Not unsurprisingly, reported rates of cancer detection and of successful treatment (used to promote particular points of view and sometimes particular treatment facilities/programs) often depend a great deal on how much cancer is actively looked for. Some of Dr. Welch's graphs (showing increased incidence of certain cancer with more screening but little change in actual mortality after more screening) were particularly illuminating.
Last edited:
