It has been known for some time that androgen deprivation therapy (ADT) may increase risk for pneumonia because, normally, androgen levels have effects on how well the immune system works and on the health of lung tissue.
According to data from a new study by a group of Canadian researchers, but based on data from the UK, prostate cancer patients on ADT are actually nearly twice as likely to be hospitalized with community-acquired pneumonia as prostate cancer patients who are not on ADT.
Now the data published by Hicks et al. do need to be treated with some caution. They come from a registry of data on men hospitalized in the UK between 1998 and 2015, and not from a prospective clinical study. However, the database is large and the study result is clearly statistically significant.
What Hicks et al. have shown is the following:
- They identified a set of 20, 310 men in the UK who were diagnosed with prostate cancer between April 1, 1998 and March 31, 2015 (a 17-year time period).
- The average (mean) follow-up period was 4.3 years
- During that follow-up period, 621/20,310 men (3.1 percent) were hospitalized with community-acquired pneumonia.
- The men who were or or had been on ADT were much more likely to be hospitalized with community-acquired pneumonia than the men who had never had ADT (hazard ratio [HR] = 1.81, i.e., an 81 percent increase in risk)
- This association was
- Normally observed within the first 6 months of treatment with ADT (HR = 1.73)
- Remained evident at ≥ 25 months on continuous ADT (HR = 1.79)
- Was a great deal lower in men who had been treated with ADT in the past (HR = 1.23)
We should note that, overall, this seems to be a risk affecting a relatively small percentage of men with prostate cancer, but clearly prostate cancer patients who are already at risk for pneumonia for other reasons need to be particularly cautious if they are started on ADT for progressive disease — and their doctors need to be aware of this too.
Filed under: Living with Prostate Cancer, Management, Risk, Treatment | Tagged: ADT, androgen, deprivation, pneumonia, risk |
THE "NEW" PROSTATE CANCER INFOLINK 
Mike:
Did they measure, or take into account 2 things: (1) How many of these men had chemotherapy, especially during pneumonia diagnosis? (2) Did they use the pneumonia vaccine in these years, and in the UK?
Thanks
Jan:
I have no specific answers for you. Remember that this was just a registry analysis of data collected on a broad basis. It was not a “study”. However, …
I think it is unlikely that any patient with active pneumonia would actually be treated with chemotherapy. If they had started chemotherapy and then got pneumonia, any scheduled dose of chemo would almost certainly be delayed until the patient’s pneumonia had resolved.
I don’t think the pneumonia vaccines would have been widely available in the UK at that time (although they might be now).
Some additional concerns about this study
Thanks for this review and the exchange involving Jan’s points. As a patient on intermittent ADT3 for 14 years, I suspect that the study may have found an association but not a cause.
First, as noted in Sitemaster’s review, the number of patients suffering pneumonia over all years of follow-up is really tiny: 621 total cases divided by 20,310 = 3.06%, ” (incidence rate: 7.2/1000 person-years)” or 0.007% of patients per each year (based on 4.3 years of follow-up). That breaks down by ADT vs. non-ADT patients to 12.1 vs 3.8 pneumonias/1000 person years, which equates to 0.012 ADT pneumonia patients per year vs. 0.004 non-ADT pneumonia patients per year. If there were a strong connection, would we not expect much higher numbers for ADT? In fact, I suspect that even a weak causal connection would lead to higher numbers.
Second, the abstract provides no information about the relative ages of the patients. I’ll bet that the ADT patients were substantially older, based on the nature of the disease as related to treatment practices. If so, pneumonia might have been relatively more frequent in the ADT group just due to age; it seems likely that age and pneumonia incidence involving hospitalization has been studied, but I have not checked. Also, older age would involve more co-morbid conditions, and that too may well be related to pneumonia incidence.
Here is my own, one-case, anecdotal evidence. After diagnosis at age 56 in late 1999, I expected to miss significant time at work because I was a cancer patient. It turned out that I rarely missed time except occasionally for medical appointments; I missed a lot less time than almost all of my colleagues. Switching to a strict Mediterranean diet and some promising supplements and keeping up exercise probably had something to do with it.
Follow-up Brief Research Supports My Second Point:
Patients on ADT tend to be older, and older people, whether or not they have prostate cancer, tend to be hospitalized more often for pneumonia, suggesting that the research finding in the subject paper is not related to ADT
I used PubMed to look into ages of ADT patients and of patients hospitalized for pneumonia. I found a number of studies for both issues, and the findings for both were consistent over the studies I checked: patients on ADT are generally older, and patients hospitalized for pneumonia are generally older.
While these trends are hardly surprising to many of us, apparently there is insufficient awareness of their existence and significance regarding ADT among some researchers.
Here are just a couple of these studies just as examples to illustrate these points:
— An Irish study re ADT and age: de Camargo Cancela M, Comber H, Sharp L. Age remains the major predictor of curative treatment non-receipt for localised prostate cancer: a population-based study. Br J Cancer. 2013;109(1):272-9. (The title means that older men are not offered curative treatment as often; instead, they often are given ADT.)
and two studies re pneumonia and age of the patient:
— Jain S, Self WH, Wunderink RD, et al. Community-acquired pneumonia requiring hospitalization among U.S. adults. N Engl J Med. 2015;373(5):415-27, which states that “…The annual incidence of pneumonia was 24.8 cases (95% confidence interval, 23.5 to 26.1) per 10,000 adults, with the highest rates among adults 65 to 79 years of age (63.0 cases per 10,000 adults) and those 80 years of age or older (164.3 cases per 10,000 adults).”
— Martín-Salvador A, Torres-Sánchez I, Sáez-Roca G, et al. Age group analysis of psychological, physical and functional deterioration in patients hospitalized for pneumonia. Arch Bronconeumol. 2015;51(10):496-501, which states that, “… Hospital admissions due to pneumonia range from 1.1 to 4 per 1,000 patients and this figure increases with age.”
A lot more detail is provided in the abstracts and studies, of course.
Pressure to Publish May Be Yielding Poor Studies Linking ADT to Various Age-Related Ailments
As a former long-term patient on ADT and IADT3 (2000-2014) prior to apparently successful curative radiation, I have paid and continue to pay attention to studies that make links between ADT and various ailments. Some of these studies are well done and help us better manage ADT. But some other studies that deal with ailments clearly related to age beg for steps to assess the proportion of the results that are due simply to age as rather than ADT.
I suspect that researchers are tempted to mine age-related ailments, such as dementia and pneumonia, for associations with long-term ADT, which is also age related. Even a spurious association with ADT may tempt an inattentive journal to publish. My prior to posts apply to the pneumonia issue.
The cost of faulty analysis in such publications is that men who would benefit from ADT/IADT may be discouraged from using what can be a highly beneficial therapy, or they and their loved ones may needlessly lose peace of mind.
We need to demand thorough control of age-driven portions of results in such studies of possible adverse effects of ADT. We also need to demand highlighting of the absolute risk of adverse effects, such as pneumonia, which can be tiny, as well as relative risk, which can be high but not consequential because of the tiny absolute risk.