Archive for August, 2013

Cancer screening redux

Monday, August 12th, 2013

The JAMA article I've mentioned in my last two posts provided an interesting sequel. It mentioned lung and bronchus cancers in the group with breast cancer and prostate cancer, those whose incidence has gone up a fair amount, and death rates have gone down.

That confused me as the numbers showed only a smidge more being diagnosed (an 8% increase) and actually a death rate increase in the same ball park (eleven precent). The footnote said a National Lung Screening Trial looking at higher risk individuals concluded low-dose CT screening could reduce death rates by 20%. I had missed this article which was published in the New England Journal of Medicine in 2011 and had over 53,000 subjects who were randomly assigned to screening with either a single chest x-ray or low-dose CT. The point was CT scanning picked up early-stage cancers and their percentage was way up, while the proportion of later-stage cancers actually was less.

Lung cancer is the third most common in both genders in the United States and is the number one cause of cancer-related deaths. If you're a heavy smoker, the chances of your dying from lung cancer go up immensely The CDC webpage on lung cancer risk factors calls cigarette smoking the major issue causing ~90% of all lung cancers in this country.

I've mentioned this before, but let me reiterate my Dad's smoking cessation and mine. He had three rooms in his private practice office and one day realized he had put a lit cigarette in the ashtray of each room, He never smoked again and lived to ninety-four. I smoked as a youngster for four years, never heavily, and I quit as a senior in medical school when I saw one of my VA patients smoking through the trachesotomy he had had for cancer.

The August edition of the Annals of Internal Medicine, published one day after the piece in JAMA, has an article titled "Screening for Lung cancer with Low-Dose Computed Tomography" (AKA a CT scan). This one reviewed four trials of CT screening; only one was large and of good quality. It actually was the research reported in the NEJM article and the project was stopped at six and a half years when it was noted how much lung cancer mortality was reduced.

What's interesting in this new review article is a series of comparisons: you need to screen 320 at-high-risk subjects by low-dose CT scanning to prevent one death from lung cancer and 219 to prevent one death from any cause. That wasn't striking until I read you need to screen 1339 women aged 50 to 59 with mammography to prevent one breast cancer death and 817 people with flexible sigmoidoscopy (a shorter version of the colonoscope that can reach an estimated two of three colon polyps) to prevent one colon cancer death.

The third category of cancer screening focused on melanoma, the worst of the three common skin cancers (the others are squamous cell and basal cell cancers). In the time frame from 1975 to 2010 the incdence of melanoma went up 199% while the death rate from this tumor went up only 32%. I found a very recent online article on this cancer, reprinted from an April, 2011 publication, by US News and World Report. It focused on its warning signs and prevention as well as treatment.

We live at  an altitude of ~5,200 feet, so our exposure to damaging  UV rays from sunlight is considerable. My wife drives 55 miles to several times a year to see a dermatologist and the last few years I've gone with her on one of those visits and had a skin check,

Try to catch these very early

Try to catch these very early

Melanomas account for three quarters of all deaths from skin cancer and are felt to result from excessive exposure to UV light. We walk in the early morning and after dark, not during the middle of the day, don't use tanning beds, my wife always wears sunscreen (I confess I don't routinely do so, but I should), and neither of us has "fair or freckled skin' or a family history of this disease.

The NCI says a change in the size or shape, color or texture of an existing mole, one of the black or brown patches that most of us, by adulthood have a number of (10-40 is common).I like the NCI's ABCDE approach to look at moles: A stand for asymmetric, different parts of the skin lesion not looking the same; B means borders that are irregular, shaggy or ill formed (uneven or scalloped); C is for color that varies from part of the mole to another; D is for diameter (melanomas, while they may be smaller, usually are larger than a pencil eraser in size): and E is for evolving (changes in size, shape, color, elevation or even bleeding, crusting/scaling or itching).

Any of these signs should prompt a quick call to your doctor and perhaps a visit to a dermatologist.

Fortunately, the only skin cancer either of us has had thus far was a squamous cell and it was removed by the Mohs technique, in which a section of the lesion is surgically removed by a specially trained physician, and then immediately stained and examined under the microscope to make sure the borders are clear of any tumor. The process is repeated until safe borders are noted.

The last kind of cancer I want to mention is thyroid; during the thirty-five year period its incidence went up 185, but its death rate went down 7%.

Be sure your doc knows if you or your family member has had a thyroid problem

Be sure your doc knows if you or your family member has had a thyroid problem

In spite of having a first cousin who had this tumor, it was the one I knew the least about. So I went to the MD Anderson Cancer Hospital's website on Thyroid Cancer Prevention and Screening. They mentioned on type that runs in families and is genetically linked, but for most of us, looking at our neck carefully twice a year, having our physician do a cancer-related exam yearly, making sure your salt has iodine, and knowing that most of these lesions occurs in women (3/4s), typically between age 20 and 55 is a good start. One other comment was exposure to radiation, especially in childhood is a significant risk factor.

There are other cancer screening procedures, but those are the common ones. Your physician may have a few more in mind.





Cancer screening Part 2: what works?

Sunday, August 11th, 2013

A blog post in The New York Times online recently addressed this issue noting, "Definition of Cancer Should Be Tightened, Scientists Say."

Here's a typical scenario: We've had a screening test for some form of the illness many of us fear the most (actually, I fear Alzheimer Disease at least as much). Then we hear the words, the really scary ones. "You've got cancer." We may not hear the modifiers that come before the C word, or anything else that follows. The National Cancer Institute's (NCI) working group wants to clarify that some conditions are pre-malignant (they're not cancer yet and, in certain cases, may never become carcinomas {another term for cancer}).

I want to return to the JAMA preprint article  I mentioned in my last post; it's titled "Overdiagnosis and Overtreatment in Cancer? An Opportunity for Improvement"

The group focused on cancer screening to include breast, lung, thyroid, colon, cervix, melanoma and prostate exams.  They tracked the incidence of a variety of cancers over the thirty-five-year period from 1975 to 2010 and found three different result patterns have been noted as more and more of us are screened for more and more conditions.

Some cancers have been diagnosed somewhat more frequently, but in many cases are less serious variants, less likely to kill us, so, overall, they are causing fewer deaths. Other malignancies are being found less frequently (due to improved and/or more frequent screening resulting in precancerous stages being treated) and the death rate from them has gone down considerably. And some cancers are being diagnosed much more often, but much of that increment is what the researchers termed "indolent disease," less active or progressive disease and overall their death rate is nearly unchanged or increased much less than the percentage change of their being diagnosed would indicate.

Let's take cancer of the prostate for an example of the first group. The incidence (rate of occurrence) in cases per 100,000 men has gone from 94 to 145.12, a fifty-four percent increase, while the death rate from the disease has gone down 30%.

So we're doing better with treating this form of cancer, seeing a lower mortality rate, but we're discovering considerably more cases that are not causing deaths. As I said in my last post, I've decided not to have any more PSA tests; at my age (72), other things are much more likely to kill me than prostate cancer. I clearly would have made the opposite choice (and did) at age 45 or 55.

Mammograms can save lives; they also can find pre-malignant lesions.

Mammograms can save lives; they also can find pre-malignant lesions.

The other tumor in the first group is breast cancer. Mammography has certainly become more of a routine procedure looking for breast lumps, but while the incidence of breast lesions has gone up 20% in the thirty-five year interval the NCI group looked at, the death rate has fallen by 30%. Much of that is due to what is termed adjuvant therapy, e.g., chemotherapy following surgery.

But another group of women with a positive mammogram have a less serious form of disease called ductal carcinoma in situ. The NCI webpage on this tumor says it is noninvasive, but some (uncertain) percentage of these can progress to an invasive form.

What is DCIS? It's the most common type of non-invasive breast cancer, starts in the milk Ducts, has been termed a Carcinoma and is In Situ which means "in its original place."

The NCI group who published the JAMA article terms it a premalignant condition and would prefer it not be called a carcinoma. Most of these lesions are small (~70% are less than 1 cm in size) and 80% are not found by breast examination, but show up on a mammogram.

Another NCI group, however, reviewed the medical literature on DCIS saying until recently the usual response to its discovery was a mastectomy. More recently there have been two randomized, controlled trials of breast-conserving surgery (lumpectomy) combined with radiation therapy and even more recently a double-blind prospective trial utilizing chemotherapy with the drug tamoxifen given daily for 5 years in addition to lumpectomy and radiation.

A control group had the surgery and radiation plus a placebo. The group that got tamoxifen did considerably better and other clinical trials are in progress.

The second group of tumors included colon cancer and cervical cancer. The statistics are fascinating. For colon cancer the incidence rate has gone down 31% and the death rate 45% and the trend is quite similar for cancer of the cervix (5 and 59% decrements)

We all have a colon, so let me focus on that cancer for a moment with a recent personal vignette.

My father had a large sessile polyp, a projecting growth without a stalk, in the very first part of his colon. It was initially benign, would have been difficult to remove and he was in his late 70s. so it was repeatedly observed and biopsied. Eventually it had a superficial layer that was cancerous. By then he was nearly ninety and the decision was to shave off layers until what was left was benign.

So that gives me a positive family history of colon cancer.

My previous colonoscopy (2006) was totally negative and without that family history my gastroenterologist would have said, "Come back in ten years."

Some colon polyps are relatively easy to remove.

Some colon polyps are relatively easy to remove.

Because of Dad's polyp, my GI doc wanted me back in seven years. And he was right; this time I had three small polyps, all of them benign and all on stalks (The technical term is pedunculated.), making them easy to snip off. I had decided to stay awake and watch and the procedure, while somewhat uncomfortable, was fascinating. I went home reasonably confident that my polyps were benign.

But there are three kinds of those colon polyps, according to the followup letter I received. Type one is of "minimal clinical significance." Type two, which I had, is benign, but potentially precancerous. Early colon cancer screening for first-degree relatives is recommended. It was suggested I speak to parent (both dead), siblings (my only sib died at fifty-seven) and children (my daughter), so they can discuss colon cancer screening with their physician.

I copied the letter to bring to my daughter on a vacation trip we're taking.

Type three is also benign, but there's considerably more chance of it turning into a cancer.

So my next colonoscopy will be in five years this time.

If I had a totally normal colon at age 72, my GI doc had said he'd not suggest any more colonoscopies ever. My routine ten-year-interval next one would have been at age eighty-two and that, he felt, is too old to screen someone whose seventy-two-year-old colon was negative for any polyps.

But with their natural history of being potentially precancerous, the discovery of types two or three leads to a change in schedule and a family search is quite reasonable.

And, I take it, that's why the incidence of colon cancer and the death rate from that malignancy have gone down over that thirty-five-year period. There's presumably been better treatments developed for those who do end up with that form of cancer, but by dong colonoscopies and removing pre-cancerous lesions (i.e., polyps), we're preventing many of them from altering into malignant tumors.

So those are examples of the group of tumors that are being discovered more often, but are causing fewer deaths and of one (colon cancer) that our screening is helping prevent.

I'll write about cancer of the cervix and of thyroid cancer and melanomas next time.




Cancer Screening Part one: Incidentalomas & PSA

Monday, August 5th, 2013

I was reading the New York Times online today and noticed an important article in the Health section. A working group from the National Cancer Institute (NCI) had just published an article in the pre-print edition of JAMA that will likely change a highly significant face of medicine for many of us.

The issue is cancer diagnosis and, in many cases, over-diagnosis. Some pre-malignant conditions, in the viewpoint of this distinguished group, now come with the word cancer attached. That may lead to extensive testing, surgery or chemotherapy (or radiation therapy) and much mental anguish (and potential physical harm) for the patient involved.

Abnormalities, potentially malignant, can be discovered while scanning or even examining for something else. Dr. Peter Libby, chief of cardiovascular medicine at Brigham and Woman's Hospital in Boston, a Harvard medical school teaching facility, wrote a June 8, 2010 piece in The New York Times titled "The 'Incidentaloma' Problem with Medical Scans." A columnist for that paper had a CT scan for other reasons; a kidney mass was detected and a three-hour operation and eventually a six-inch scar ensued, yet the mass was benign. Dr. Libby's review of the medical literature with his area of expertise in mind showed that greater than eight percent of cardiovascular imaging studies revealed incidental findings that led to further medical procedures.

His conclusion was we're doing far too many CT scans.

Another physician wrote an April, 2011 piece in US News and World Report about a woman referred to him as a followup of an ER visit for abdominal pain that turned out to be viral gastritis. She too had a CT scan which showed her liver and intestines were normal, but one of her kidneys had a tiny mass, almost certainly a benign cyst. But the radiologist, while noting this lesion had all the features of something non-cancerous, covered his or her behind by saying, "Cannot rule out malignancy. Clinical correlation required." Translation: it was almost certainly nothing serious, but there was a very small chance that it might be cancer, and now it was the surgeon's job to make sure it wasn't.

But it's not just those advanced radiologic procedures, or MRIs or other tests; It's the mentality involved and that includes physical exams.

Let me give you a personal vignette. In 1969, as a second year clinical fellow in Nephrology, I went to see the Chief of Urology because of an abnormal lab test that involved my kidney function. A few questions later, he determined it was due to a diet I was on for a research project.

"But as long as you're here, Peter," he said, "I'll check your prostate."

It wasn't quite normal, a tad bigger than it should have been. He said, "You'll have a TURP (transurtheral resection of the prostate) by the time you're sixty"

He recommended I monitor my symptoms (I had none at the time) and get repeat examinations at intervals.

Well, I'm seventy-two and haven't had that surgical procedure. But what I did have, for many years, was a yearly digital prostate exam and, after 1994, a PSA (prostate specific antigen) blood test done periodically. At some point not only was the gland enlarged, but it the urologist involved thought it was also slightly asymmetric, so I had multiple biopsies, even though my PSA had consistently been less than 1.0, e.g., way below the level of concern.

All those biopsies showed benign (non-cancerous) tissue.

Now that I'm over seventy I don't ask for a PSA test, wouldn't agree with one if it were suggested and had my last digital exam of the gland several years ago.

But if I had that cancer at age 28 or 40, I would have been in real trouble. I would have been concerned about a malignancy that would likely kill me and would have welcomed any logical treatment for the disease. The NCI webpage on prostate cancer estimates nearly 240,000 new cases will be diagnosed in the United States this year and almost 30,000 men will die from that disease.

But the natural history of the tumor in most older men (>70 years old) is such that they will most likely die from something else (e.g., heart disease).

The NCI's fact sheet on the PSA test is well worth reading. An initial statement is that the higher a man's PSA level is, the more likely he has cancer of the prostate. Then the caveats begin: there are other reasons for the PSA to be elevated (I now have one of those, BPH or benign prostatic hypertrophy, an enlarged prostate). Half of all men over the age of fifty have BPH that is symptomatic with some hesitancy in starting their urine stream and/or a smaller stream.

If I had actually had cancer of the prostate at age sixty, my PSA may have been elevated. Then I would likely have had surgery to remove the tumor if it was localized to the gland and my physicians would then have periodically repeated my PSA testing to monitor if I had a recurrence of the neoplasm (a new and abnormal growth of tissue in some part of the body, especially as a characteristic of cancer).

A PSA level under 4.0 (nanograms/mililiter) was considered to be normal (my last PSA, done three years ago, was 0.7 ng/mL). In the past, if a man had a PSA level above 4.0 his physician  would have likely suggested biopsy to see if he had prostate cancer. But there are men who have that malignancy and yet have a PSA under 4.0 and, as I mentioned above, other, benign conditions can elevate the PSAwhile some medications used for BPH can lead to a lower PSA level. Only a quarter of those guys who have a prostate biopsy because of a PSA that's elevated actually have cancer of the gland.

But now there's considerable question whether PSA testing should be a routine, even though medicare and many private medical insurance plans cover a yearly screening using this test. The consensus seems to be that men should hear the pros and cons of the test before giving consent for it to be performed.

If 1,000 men in the 55 to 69 age range get screened with this test every one to four years, 100-200 will have false-positive results (no cancer, but an elevated PSA) and may have a biopsy recommended and, of course, worry about what's going on with their prostate.

One hundred ten, according to the website, will be diagnosed as having prostate cancer and nearly half of those will have treatment complications (The National Cancer Institute website mentions sexual dysfunction, bowel or bladder control issues and infections.)

Four to five will actually die from prostate cancer, but five of every 1,000 who don't have the screening will die from that cancer.

So the net is 1,000 have been screened to save 0-1 life.

And to further complicate things, the research done to determine what the normal upper limit of PSA is has largely been done in white men only.

So where do we go from here? There are studies being done to look at precursors of PSA, rate of change of PSA. free versus bound PSA etc.

We need  a better method to tell us if a man has a cancer or a benign prostate condition and to determine which prostate cancers are highly malignant fromt hose that are slow growing.

And all that's just one cancer-screening tool.