Is CLL an “Old Man’s Disease”?

Old ageOne of the often heard clichés is that CLL is an “old man’s disease”. As with all clichés, there is a smidgen of truth in it. A 2008 article  in “Blood” looked at CLL patient profile in USA over the time period of 1973 – 2004.

  • The single most common age group diagnosed with CLL was 70 -79 years old. Roughly 30% of the cases registered fell into this age bracket.
  • 60% of the cases were men, a statistical difference that persisted over the whole time period of the study.
  • However, another way of looking at the same statistics is that a goodly percentage of the patients were younger than 70 years and fully 40% were women. And I am dead certain all of the patients were young at heart. So there.
  • Comparing the time frames 1980-84 and 2000-04, the number of registered CLL cases increased by 20%. But before you jump to conclusions, the authors point out that the number of CBC blood tests done as part of routine annual checkups has increased dramatically over the two decades and this may be the reason why we are now seeing more cases of CLL that would have gone undiagnosed in the earlier time frame. It is a case of finding more of them when you go looking for problems.
  • Even with the higher level of routine monitoring of blood counts, the authors believe CLL is an under-reported cancer. The actual number of cases may be higher than reported by the national registry SEER. This may be due to the fact that majority of CLL patients are not treated at diagnosis. Local oncologists sending in their data to SEER registries may overlook counting patients that are in Watch & Wait mode.

Age at diagnosis SEER

As you can see from the table above, fully 20% of the patients diagnosed with CLL over this time period (1980 through 2004) were young people in the age range of 15-59. These guys are mere spring chickens as it were, still busy building careers, raising kids and hopefully saving money for a distant retirement.

That raises an interesting question: does CLL have a different face in younger patients? Is age at the time of diagnosis an important factor? Should we be more aggressive in doing all the prognostic testing in the case of younger patients? A Mayo Clinic abstract (below) at the recent ASH2009 conference addressed this question.

Influence of Age at Diagnosis On Utility of Prognostic Testing in Patients with CLL

Tait D. Shanafelt, M.D., Kari Rabe*, Neil E. Kay, MD, Clive S. Zent, MD, Diane F. Jelinek, PhD, Susan Schwager*, Deborah Bowen*, Susan L Slager, PhD, Megan Reinalda*, Curtis A. Hanson, MD and Timothy G Call, MD

Mayo Clinic, Rochester, MN

PURPOSE: Numerous clinical and biologic parameters are able to predict time to first treatment (TFT) and overall survival (OS) for patients with CLL. The utility of prognostic testing in CLL patients may vary based on age at diagnosis given the higher mortality from other causes among older individuals. The goals of the present study were to i) evaluate differences in natural history of CLL patients based on age at diagnosis, ii) compare survival to age-matched individuals in the general population, and iii) determine the age-stratified utility of prognostic testing.


All 2487 CLL/SLL patients diagnosed between January 1995 and June 2008 and cared for in the Mayo Clinic Division of Hematology were categorized by age at diagnosis (<55 years, 55-64 years, 65-74 years, or ≥75 years) and evaluated for differences in clinical characteristics, TFT, and OS using chi-square tests. IGHV gene mutation analysis, ZAP-70 status, CD38 status, and cytogenetics abnormalities by interphase FISH testing were performed as part of clinical and/or research studies using methods previously described by our group. Cox regression was used to evaluate the ability of prognostic parameters to predict clinical outcome stratified by age at diagnosis. Patient survival was compared to that of the age matched Minnesota population.

RESULTS: When grouped by age, 593 patients were age <55, 713 age 55-64, 748 age 65-74, and 433 age ≥75 at diagnosis. As of last follow-up, 268, 282, 296, and 108 patients in these age groups have been treated, respectively and 112, 171, 254, and 199 patients have died. Survival decreased as the age at diagnosis increased (p<0.001). Overall Survival of CLL patients was significantly shorter than that of the age-matched general population for patients age <55 years (p<0.001), 55-64 years (p<0.001), and 65-74 years (p<0.001) at diagnosis but not those ≥75 years (p=0.136). Among Rai stage 0 patients, survival was also shorter than that of the age-matched general population for CLL patients age <55 years (p<0.001), 55-64 years (p<0.001), and 65-74 years (p<0.001) but not those ≥ age 75 at diagnosis (p=0.07).

No statistically significant differences were observed in CD38 status, IGHV mutation status, ZAP-70 status or the frequency of common cytogenetic abnormalities on FISH based on age at diagnosis category. Consistent with prior reports, all of these parameters were powerful predictors of TFT and OS for the overall cohort on univariate analysis (all p<0.001 for both TFT and OS). Given the variation in TFT and OS based on age at diagnosis, we next evaluated the utility of these prognostic parameters for predicting TFT and OS in each age category after adjusting for stage. CD38, IGHV mutation, and ZAP-70 each predicted TFT independent of stage for all age groups (all p <0.05). FISH results predicted TFT independent of stage for patients age 55-64 and 65-74 however did not reach statistical significance for those age <55 (p=0.18) or ≥75 (p=0.07). The utility of these parameters for predicting OS independent of stage was less consistent and varied by age. IGHV predicted OS independent of stage for patients <age 55 (p<0.001), 55-64 (p=0.007), and 65-74 (p≤0.001) but was not a significant predictor of OS among those ≥75 (p=0.15). CD38, ZAP-70, and FISH each predicted survival independent of stage for only 2 of the 4 age categories (ZAP-70 for age <55 and 55-64; FISH for age <55 and 65-74; CD38 for 55-64 and ≥75).

CONCLUSIONS: Survival of CLL patients is shorter than that of the age-matched general population for individuals age <75 at diagnosis regardless of disease stage. In contrast, survival does not differ from the age-matched general population among CLL patients ≥age 75 at diagnosis. Prognostic testing using CD38, IGHV mutation, and ZAP-70 is useful for predicting TFT independent of stage for CLL patients of all ages (including those ≥age 75) but has less value for predicting OS particularly as the age at diagnosis increases. IGHV appeared to be the best predictor of OS independent of stage among all patients age <75. Unexpectedly, prognostic testing had little utility for predicting OS independent of stage among patients age ≥75. These findings have important implications regarding the use of prognostic tests in patients with CLL as well as the use of test results to select patients for clinical trials testing early treatment for asymptomatic individuals.

What These Statistics Mean for Your Life

Sometimes my eyes glaze over when I look at a block of writing like this and it is hard to get the punch-lines. Let me see if I can make it a little easier on your “aging” eyes. The chart below describes age distribution of the patients seen at Mayo Clinic over a 12 ½ year period.

Age at Diagnosis, death

As you would expect, just over 23% of patients were younger than 55 years when they were diagnosed – roughly the same as the country wide statistics reported by SEER. Bulk of the patients were in the three older age groups.

Patient Deaths Over Study Period

Mayo reports the number of patient deaths in each age group over this study period. The trend is clear. Older patients were more likely to have died during the study period. “Only” 18.9% of the youngest group died but a whopping 46% of patients who were over 75 years at diagnosis died during the study period.

Does this mean older patients pay a higher penalty of death due to CLL compared to younger folks? No!! When compared to a similar age cohort who did not have CLL, there was no statistical difference in the percentage of deaths in the over 75 group. In other words, CLL did not have much of an impact on overall survival for this oldest group.

It makes sense when you think about it. By the time people get to be 75 years old, they tend to accumulate a bunch of other health issues. Twelve years later, they are pushing 87 years and it is reasonable that slightly less than half (46%) of them are dead by then. For this group of patients CLL is just one more health issue to deal with, along with all the other issues such as heart disease, brittle bones, diabetes, other cancer etc.

Moral of the story seems to be this: if you are 75 years or older at the time of diagnosis of CLL, it is most likely not going to be a big deal in the sense of reduced life expectancy. If you are also blessed with a relatively smoldering variety of CLL, you may want to consider not doing anything too aggressive by way of therapy. Chances are good you will be able to “run out the clock” with kinder and gentler therapy interventions to hold the line as necessary and die peacefully in your bed on account of something other than CLL.

The game plan for you might be looking to maximize the quality of your life, enjoying each and every day that you have coming to you, keeping a wary eye out for all the other things that can go wrong as you age. All of us have to go someday, and chances are good you would have lived out your natural life span before CLL gets you.

Please bear in mind, I am talking in generalities here. Each individual is different and has different expectations. You might come from a family that typically live into their late nineties, in which case you would not be happy about kicking the bucket when you are several years short of that goal! I have a very young at heart 90+ year old friend who has no intention of dying – ever, if he can help it. On the other hand, my mom at 83 thinks she is ancient and chomping at the bit to get done with it. It is all a matter of your personal expectations, your zest for life.

Bitter Pill for Younger Patients

The story is dramatically different at the other end of the age spectrum. As the table shows, younger patients paid a huge penalty in terms of their reduced life expectancy.  Compared to similar age cohorts they died at a younger age and this difference was statistically significant. Young guys who are 55 years at diagnosis would be only 67 years old 12 years later. You would not expect a fifth of 67 year olds (18.9%) to die! Heck, most of them would have just become eligible for Social Security and hitting the golf courses, but for CLL. Telling these guys CLL is a “good cancer” is adding insult to injury.

“Managing” this incurable cancer requires savvy gamesmanship in younger patients. For starters, they have a lot more responsibilities to juggle as they struggle with CLL.  Jobs, money, medical insurance, taking care of younger children – all of these are bigger issues when you are in the prime of your life.  Dealing with CLL on top of everything else can get to be a real challenge.

Fortunately younger patients are also in better shape to handle more aggressive therapy since they are generally healthier and have fewer secondary health complications. Chances are good that they will not be able to nickel and dime this disease with low impact therapy and thereby run out the clock and live their normal life expectancy. For very young patients, the million dollar question is this: when (not if) to go into a mini-allo stem cell transplant. To date, this is the only technology known to CURE CLL. Every other therapy out there is able to do no more than give you breathing room – while you get your ducks in a row for an allogeneic transplant. The other choice is dying a lot sooner than you would have without the CLL.  Sometimes it is a case of deciding between the devil and the deep blue sea.  The good news is that transplant statistics are improving each year, the devil is not so scary anymore and far preferable to the deep blue sea.

Prognostic Testing

Over the last 10 years or so, we have learned a great deal about what makes CLL tick, what to look for by way of prognostic indicators. Most experts agree that top four prognostic indicators are:

  • IgVH gene mutation status
  • FISH test (chromosomal abnormalities)
  • CD38 level on CLL cells (Flow cytometry)
  • ZAP70 level on CLL cells

If you are not familiar with these prognostic indicators and how they influence your “Risk Bucket”, I strongly urge you to read an article “What type of CLL do you have” on the Newly Diagnosed page.

The Mayo study found that prognostic test results were useful in predicting the time to first treatment in all age groups. But they had less success in predicting overall survival – especially in older patients, those over 75 at the time of diagnosis. By the time patients are 75 or older, other medical issues can become more important, their CLL may no longer be in charge of driving the bus as it were.

Of the lot, IgVH gene mutation status was the single best predictor of overall survival for patients younger than 75 at diagnosis. This is also the test that local oncologists seem to balk at prescribing, possibly because it is harder to understand and interpret.  If your guy does not “get it”, you have two choices.  Go to an expert center where they do get it.  Or, learn as much as you can about the test, arm yourself  with reputable journal articles on the subject and try to convince your oncologist why getting this test done is important.

Moral of the story: prognostic testing is a lot more important in younger patients. Money and effort spent in getting prognostic tests done are likely to pay off in giving them a better awareness of what they are likely to see coming at them down the road. They have a more complex game plan to implement and any heads-up they can get will be invaluable.

Israeli Study

Below is a very recent paper looking at some of the same issues. This study comes to us from Israel. It is a much smaller study than the Mayo report we discussed above, only 87 patients participated in it.

This paper also documents the larger penalty of loss of life expectancy paid by younger patients. Unlike the Mayo study, this paper suggests that younger patients are likely to have a somewhat more aggressive form of CLL, compared to patients who are older at diagnosis. I cannot swear to its statistical significance, but my own strictly anecdotal database supports this. Over the last seven plus years I have seen many more young patients with more aggressive CLL. Very few of these young guys seem to have classic “smoldering” CLL variety. On the other hand, I have dozens of older patients in my database that have lived 15 or more years without needing any therapy at all for their CLL!

Int J Clin Pract. 2009 Nov;63(11):1601-3.

Evaluating the impact of age and disease on survival of chronic lymphocytic leukaemia patients by a new method.

Klepfish A, Rachmilewitz EA, Sarid M, Schattner A.

Blood Bank and Hematology Institute, Wolfson Medical Center, Holon, Israel.

BACKGROUND: The impact of chronic lymphocytic leukaemia (CLL) on survival may be different in younger patients, but this remains controversial. OBJECTIVES: The aim of the study was to examine the effect of age on survival in CLL using an original method. METHODS: Clinical, laboratory and survival data of 87 CLL patients treated in our institute were analyzed. The survival of patients in different age groups was determined and compared, as related to the expected survival of age- and gender-matched general population obtained from national statistical data. RESULTS: The mean age in the younger (< or = 65 years, n = 37) and older (> 65 years) age groups was 56 and 74 years (p < 0.001). The younger group had more unfavourable presentation, with advanced stage (Rai 2-4) in 46% vs. 16% (p = 0.002), and diffuse involvement of bone marrow in 60% vs. 18% (p = 0.03), compared with the older group, and were more likely to require treatment (p = 0.02). The Kaplan-Meyer curve showed a more favourable survival for the younger group. However, the loss of expected survival exposed a reversed pattern: while the older patients lost only 13%, the survival loss in the younger patients was 44% (p < 0.001). CONCLUSIONS: Chronic lymphocytic leukaemia had a more unfavourable presentation and a more severe clinical course in the younger patients. Our method of evaluating the negative impact of disease on expected survival reveals that their survival also is significantly more affected than that of older patients. We suggest calculating the loss of expected survival as a new criterion for assessing disease impact.

PMID: 19832816

All of us are unique individuals with our own very personal take on things as important as life and death.  No amount of statistical data is able to describe our individual zest for life, our willingness to fight one more battle today so that we can see the next tomorrow.  Cultural differences are often over looked by physicians.  Modern western culture tends to favor ever more intervention, doing anything and everything to prolong life. Not so in some Eastern cultures where fatalism plays a big role in accepting death more readily.  Perhaps having fewer healthcare dollars to spend has something to do with it!?  What say you?

Poppies stained glass