Cancer is ubiquitous and one of the leading causes of death in both so called first world and developing countries. The commonest non-cutaneous malignancy in men is of the prostate with a lifetime cumulative risk of 9%.This translates to 1 in 6 men getting prostate cancer (PCa) at some time in their life.
COMMENTARY FROM AN AFRICAN PERSPECTIVE
Professor Shingai B A Mutambirwa
MBChB (Zim), M Med (Urol) (MEDUNSA), FCS (Urol) SA
Head: Sefako Makgato Health Sciences University
Cancer is ubiquitous and one of the leading causes of death in both so called first world and developing countries. The commonest non-cutaneous malignancy in men is of the prostate with a lifetime cumulative risk of 9%.This translates to 1 in 6 men getting prostate cancer (PCa) at some time in their life. In addition PCa is the 2nd most lethal malignancy after lung cancer. Although prevention of cancer is the ideal the few lifestyle ,medical and environmental factors that have been targeted have met with a varying and generally disappointing degree of effectiveness. Furthermore due to our inability to accurately predict or prevent cancer it is generally acknowledged that cure if at all possible for cancer is best accomplished by early detection to decide on the most appropriate and efficient treatment. PCa theoretically should be no exception to this rule and screening for PCa using the aids of a digital rectal examination and a blood test for the trypsin-like enzyme prostate specific antigen (PSA) has been the mainstay of PCa screening for over 20 years in first world countries with what appeared to be great success. However controversy over PCa screening has abounded since the press releases early in 2013 that this notion might be flawed with quotes in major mainstream publications such as; “PSA testing saves few lives and leads to risky and unnecessary treatments for large numbers of men“- USA Today and “Urology group stops recommending routine PSA Test” – Washington Post. These commentaries were triggered by a number of studies on PCa screening and treatment which were released in 2012 and 2013 and that were analysed by the United States Preventative Services Task Force (USPSTF) an astute body set up to oversee general issues affecting American public health care. USPSTF has produced many useful recommendations and is a relevant regulator for the USA. Therefore it is of no surprise that these recommendations from USPSTF have led to an almost 50% drop in PCa screening in the USA over the last year. However USPSTF’s statements have been challenged not only by the American Urological Association (AUA) (-9) but many other associated bodies including the South African Urological Association (SAUA) and the Prostate Cancer Foundation of SA (PCF) for a number of reasons which this commentary wishes to address.
First it must be understood that the USPSTF recommendations are for the USA and cannot be directly extrapolated to any other nation or population group. As a prime example of this heterogeneity of PCa in different countries is the issue of stage migration (SM). SM in cancer refers to the fact that as screening occurs more low stage/risk cancers (LR) are identified and treated whilst high stage /risk metastatic cancers (HR) decrease proportionately. With >20 years of widespread PSA based screening for PCa by American health care providers this has occurred in the US where upwards of 90% of PCa are diagnosed when LR. This is unlike in most African countries where HR and metastatic PCa predominates. Therefore it stands to reason that in Africa there is still a need to identity LR in our population to get this same SM. In fact the same data the USPSTF has used to make the recommendations on PCa screening in American men actually show this to be the case(-2). Besides the SM issue there are other issues of contention with the USPTF’s analysis particularly for countries that have not had a history of PCa screening. The aim of the rest of this commentary is to expand on how USPTF got to their conclusions and critique the analysis to give a view for PCa screening recommendations in Africa.
First we will assess the background to the USPTF’s conclusion that PCa screening is obsolete i.e. that early diagnosis is unnecessary and does not provide an advantage for PCa survival. Secondly we will address the conclusion that treatment of PCa is ineffective and has too many side effects (S/E). To address the former it will have to be shown if there is sufficient data on whether PSA screening saves lives. For the latter we will take a look at the trials on radical surgery for potentially curable PCa versus watchful waiting for PCa which have prompted the controversy. 1-Does PSA screening save lives? At the outset it must be pointed out that there is a difference between public health screening (PHS) and individualized patient care screening (CS) as confusing the 2 has led to much of the confusion about this issue. PHS refers to screening populations mandated by medical policy in a country extending to all patients at risk of the target of disease.An example in many African countries could be in HIV. Unfortunately in the context of cancer PHS can often be difficult if not impossible to implement particularly due to the expense of such an exercise. PHC for PCa has therefore never been the recommendation or policy of any country including the US. CS on the other hand refers to testing individual patients who voluntarily consult their health care provider for whatever reason the patient deems necessary. As alluded to above widespread PSA testing started in the 1990s in the US but this was not PHS but rather a form of accelerated CS driven by new knowledge on the part of health care providers. And it changed the landscape of PCa in that country. Data from the early ‘90s showed 68% of patients in the US had LR whilst 21% presented with bone metastases.This initiated the concern that earlier diagnosis of PCa needed to be addressed. As the PSA testing drive progressed data from 2009 showed these figures had changed with 91% localised disease , 5% with locally advanced or visceral/lymphatic spread and only 4% of patients with bone metastases at presentation. This is significant as it is recognised that once PCa is metastatic it is inevitably fatal even though androgen deprivation therapy can be palliative for an extended period ,albeit often with significant S/E, due to the fact that essentially all metastatic PCa will eventually become castrate resistant. Therefore if a very modest age – adjusted death incidence of PCa of 39/100 000 in 1990 is applied had there been no PSA testing or improved treatment there would have been 59 000 death from PCa that year in the USA alone.
With PSA testing being performed on the majority of men by 2007 even with the increase in the US male population the mortality from this disease had dropped to 35 000 deaths and in 2013 this had further reduced to 29 700. Despite these compelling data the problem the USPTF had were the results of a large randomized trial for screening of Prostate, Lung, Colorectal and Ovarian Cancer (PLCO) which showed some benefit in screening for all these cancers except prostate in the US. However in the same journal (-6) a similar study conducted in Europe the European Randomized Study of Screening for Prostate Cancer (ERSPC) trial was published which in contradiction to PLCO showed a 20% reduction in deaths from PCa with PSA screening though admittedly to achieve this positive effect 1000 men had to be screened to prevent one death. Regardless of the contradictory result two of the problems with both were that the major endpoint was mortality at 10 years and the only S/Es considered were incontinence, erectile dysfunction and rectal injury. 10 year mortality was used as the end point despite the urological community having known for years after numerous well conducted trials (-1; -2) that this length of follow up is inadequate.It has clearly been shown that the vast majority of men who die from PCa within 10 years of diagnosis after screening did not have curable disease at diagnosis and therefore would not have benefited from the screening. Therefore screening for PCa in the scenario of patients with limited lifespan and/or comorbidities not projecting a life expectancy >10 years has again never been advocated by the AUA (-9) or any other medical body. In the same vein most men with curable disease detected by screening even if left untreated generally do not die from their PCa within 10 years of diagnosis. Mortality from the disease only rises after the first decade of diagnosis. It therefore stands to reason that if a man has a >10 year life expectancy PSA screening should be done as the majority of these PCa if left undetected and therefore untreated will cause mortality to climb exponentially between 15 – 20 years.
Regarding screening in younger men it has been shown that a PSA level below 0.7ng/ml under 50 years of age or 0.9ng/ml above this age is predictive of a significantly decreased probability of HR in the future(-3). Therefore if a baseline PSA is above this threshold more stringent PSA testing is warranted whilst if below this level screening intervals can be deferred to longer intervals. These data do not seem to be addressed by the USPTF and particularly for black African males there is accumulating evidence that they are at higher risk for PCa,present at a younger age and possibly have worse outcomes stage for stage. At the other end of the age spectrum the USPSTF does not recommend screening for men over the age of 70.However we know that 50% deaths from PCa occur in men diagnosed after 75 (-4).Therefore this recommendation needs to be adapted to each nation’s needs as the aging population grows. Of further concern for countries deciding to stop PC is that it has been shown that 9 years after stopping CS the incidence of potentially lethal PCa equalises to the incidence in a non-screened population(-5).
2-Does treatment really decrease PCa mortality and complications?
To address this the USPSTF mainly considered data from the Prostate Cancer Intervention Versus Observation trial (PIVOT) (-6). PIVOT did not show benefit for outcomes of the gold standard for treatment with curative intent in LR which is radical prostatectomy (RP) when compared to observation until progression. Unfortunately from detailed analysis of PIVOT it appears that in the study population older sicker men who from the foregoing data should have been observed were randomised to the surgery arm whilst healthy men who were candidates for surgery were not given the surgery option. This is reflected in the fact that although PIVOT planned for accrual of 2000 healthy men with at least a 10 year lifespan it actually recruited only 731 men and of these at 10 years 50% were dead from other causes. To compound this at 14 years only 30% of any of the cohort was still alive. Despite this PIVOT still showed a 60% decrease in PCa metastases and a 40% reduction in PCa deaths in men with a PSA level >10ng/ml at baseline who were treated with RP. To emphasize this point data from a similar study conducted in the pre PSA era by the Scandinavian group of Lars Holmberg and Anna Bill–Axelson (-7) over a more representative 15 year follow up showed there were significant decreases in metastases, death from any cause and death from PCa in all PCa groups irrespective of age and including men with LR when treated with RP.As to the data alluded to above that the only S/Es the PLCO & ERSPC trials mention are death, erectile dysfunction, incontinence and rectal injury it has to be said that those these S/Es are of significant concern and were higher in the RP arm. However there is no mention or consideration in PIVOT of the S/Es from hormonal therapy, metastases pain etc. that precede death from fatal PCa. To the patient and treating physician the latter are often their biggest worry and are proven to occur more frequently in untreated patients (-8).
So what should the average African male and his health care provider do with these data?
Firstly to balance the argument it must be recognised that screening with PSA does lead to a significant amount of over diagnosis which may lead to over treatment of insignificant PCa possibly by as much as 60%. However in the current environment in Africa at least until better PCa markers, stratification and less morbid treatments are available we have little choice but to still use PSA for screening in the CS situation. Secondly for the medical system of any country there is currently no place either fiscally or from medical outcomes data to justify PHS for PCa. Thirdly if a man over 40 years has a limited lifespan (less than 10 years) for whatever reason CS should be used cautiously or not at all (-10). In all other men with a life expectancy >10 years there should be the option to have CS for PCa after discussion with his physician on the pro and cons of screening for the disease including the fact that diagnosis will usually require a prostate biopsy if his PSA and/or digital rectal examination is abnormal. If PCa is diagnosed treatment options will need to be explained to the patient and individualised to his risk and choices and if LR is identified it must be stressed that cure can be achieved in the majority of patients. Furthermore active surveillance is now a recognised option which may avoid potentially morbid curative treatments although follow up protocols still need be streamlined especially in light of the emerging data of possibly poorer outcomes with this strategy in African American men.
Finally it must be emphasised that the most important person in the decision making process is the patient who must be fully informed of his rights and the possible outcomes of any screening or procedures that may be necessary and be holistically involved throughout the whole process.
1. Etzioni et al: Cancer 2012: The Prostate Cancer Conundrum Revisited
2. Pipiolek et al: Natural History of Early, Localized Prostate Cancer:A final report from Three Decades of Follow up
3. Vickers et al: BMJ 2013. 346
4. Scosyev et al – Cancer 2012 – 118; 3062
5. Berdahl et al – European Urology – Epub May 8 years after Spoppune 2013
6. PIVOT Trial NEJM July 19 2012 Vol 367 No3
7. Lars. Holmberg and Anna Bill – Axelson NEJM May 5 2011
8. Hartzband & Groopman NEJM Sept 13 2012 There is more to life than Death
9. AUA Guidelines: Early detection of Prostate Cancer: J Urol 2013 – Epub 190
10. Carlsson et al J Clin Oncol 2012 30: 2581 – 84
Extended References available on request