The ban on phenacetin is associated with changes in the incidence trends of upper-urinary tract cancers in Australia

The ban on phenacetin is associated with changes
in the incidence trends of upper-urinary tract
cancers in Australia
Sebastien Antoni,1 Isabelle Soerjomataram,1 Suzanne Moore,1 Jacques Ferlay,1 Freddy Sitas,2-4
David P. Smith,2,5 David Forman1
Abstract
Background: Phenacetin is an analgesic that causes renal diseases and cancers of the upper-urinary
tract (UUT). It was banned in most countries from the late 1960s. This study aimed to
evaluate, for the first time, the long-term population impact of the phenacetin ban on UUT
cancer rates.
Methods: We used cancer registry data from Australia, where phenacetin was widely used,
to study age- and sex-specific incidence trends of cancers of the renal pelvis and the ureter
after the phenacetin ban (1979). Incidence rate ratios and average annual percentage change
(AAPC) were calculated to quantify changes in rates over time.
Results: Incidence rates of renal pelvis cancer decreased by 52% in women and 39% in men
between 1983–1987 and 2003–2007. The decline in women was stronger in states where the
use of phenacetin was the most widespread, e.g. New South Wales (AAPC: -4.1%; 95% CI -5.3,
-2.9) and Queensland (AAPC: -3.3%; 95% CI -4.9, -1.8), and after the mid-1990s. Incidence rates
of ureteral cancer remained stable for both sexes throughout the study period.
Conclusions: Our findings strongly suggest a beneficial impact of the ban on phenacetin on
the incidence of renal pelvis cancer in Australia, particularly among women.
Key words: phenacetin, analgesic, renal pelvis, neoplasm, Australia
2014 Online Australian and New Zealand Journal of Public Health 1
© 2014 Public Health Association of Australia
Analgesics had been hypothesised to
cause renal disorders since the early
1900s. However, their long-term side
effects were only identified in the 1950s when
their association with a kidney disease called
analgesic nephropathy was demonstrated.1
This association was specifically attributed
to phenacetin, a compound widely used in
analgesics at the time. This led to a legal ban
on phenacetin in most countries from the late
1960s.1,2 An increased risk of upper-urinary
tract (UUT) cancers, correlated with the level
of exposure, had also been reported among
regular users of phenacetin-containing
analgesics.3,4 This resulted in phenacetin
being classified as carcinogenic to humans by
the IARC Monographs program.5,6
Australia was one of the countries where
analgesic mixtures containing phenacetin
were popular. In the 1950-60s, they were
heavily marketed through advertisements
such as the famous slogan, “Have a cup of
tea, a BEX and a good lie down”, which often
targeted women. Such marketing resulted in
a high prevalence of use of these analgesics
and later to high incidence rates of analgesic
nephropathy, particularly in women.7 In
the mid-1970s, papillary necrosis, the main
feature of analgesic nephropathy, became
the second most common cause of end-stage
renal disease and accounted for 15-20% of
patients entering dialysis and transplant
programs in Australia.7 Following the
phenacetin ban in 1979, incidence rates of
analgesic nephropathy decreased.8 However,
little is known on the impact of this public
health prevention measure on UUT cancer
incidence.
To date, only one study in the Australian State
of New South Wales investigated UUT cancer
trends following the ban on phenacetin.9 It
reported a slight, non-significant decrease
of UUT cancer rates between 1985 and 1995.
Although suggestive that the phenacetin ban
had a beneficial effect on UUT cancer rates,
the study was limited by the length of the
time series available. To follow up on these
observations and to determine the long-term
population impact of the ban on phenacetin
on UUT cancer rates, we evaluated the
incidence trends of UUT cancers in five
Australian States over a 25-year period
(1983–2007) following the phenacetin ban.
Methods
Incidence data were obtained from the
population-based cancer registries of New
South Wales, Queensland, South Australia,
Victoria and Western Australia, which cover
about 95% of the Australian population.
Other states did not provide sufficient
numbers of cases for analyses. The upper-urinary
tract has two paired organs, the
renal pelvis and the ureters. We therefore
included all incident cases of transitional
1. Section of Cancer Surveillance, International Agency for Research on Cancer, Lyon, France
2. Cancer Council NSW, New South Wales
3. School of Public Health, University of Sydney, New South Wales
4. School of Public Health and Community Medicine, University of New South Wales
5. Griffith Health Institute, Griffith University, Queensland
Correspondence to: Mr Sebastien Antoni, Section of Cancer Surveillance, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon Cedex 08,
France; e-mail: antonis@iarc.fr
Submitted: November 2013; Revision requested: February 2014; Accepted: March 2014
The authors have stated they have no conflict of interest.
Aust NZ J Public Health. 2014; Online; doi: 10.1111/1753-6405.12252

Figure 1: Trends in renal pelvis (A) and ureteral (B) cancer incidence in Australian men (straight line) and women
(dashed line) aged 30-79 between 1983 and 2007. Rates were smoothed using a locally weighted regression and
are presented on a logarithmic scale.
Renal pelvis
Men Women
A
1980 1985 1990 1995 2000 2005 2010
Calendar year
Age−standardised incidence rate
(per 100,000 person−years)
0.1 0.2 0.5 1.0 2.0
Ureter
2 Australian and New Zealand Journal of Public Health 2014 Online
Antoni et al.
cell carcinomas (ICDO-3 morphology codes
8120-8131) of the renal pelvis (ICDO-3
topography code C65; n=4,387; 1,979 men
and 2,408 women) and of the ureter (ICDO-3
C66; n=1,532; 924 men and 608 women) in
persons aged 30–79 and diagnosed between
1983 (when complete national registration
became available in Australia) and 2007.
Official population data were obtained from
the website of the Australian Bureau of
Statistics.10
Age-standardised incidence rates (ASR) were
computed separately for renal pelvis and
ureteral cancers using the world standard
population.11 Incidence rate ratios (IRR) and
their 95% confidence intervals (CI) were
calculated to compare incidence rates
between sexes or between time periods.12 For
these analyses, we used two five-year periods
(1983–1987 and 2003–2007) to obtain more
stable results. To quantify the annual variation
in the ASR over a period of time, we estimated
annual percent change (APC) and their 95%
CI by fitting a simple linear model.13 When a
significant change occurs in incidence trends,
an APC estimate is obtained for the periods
preceding and following such change. Thus,
to assess the variation in the ASR over the
whole study period (whether or not changes
occurred in the trends), we also estimated
Average Annual Percent Change (AAPC) and
their 95% CI.13 AAPC are weighted averages
of the different APC values, the weights being
the durations of the different intervals. APC
and AAPC were computed using Joinpoint
regression analyses.14 We performed age-period
and age-cohort analyses to investigate
the effects of the year of diagnosis and of the
birth cohort on the observed trends.15 Plotted
incidence curves were smoothed using locally
weighted regression.16 All analyses were
performed using R v3.0.1 & JoinPoint v4.0.1.
Results
In 1983–1987, the highest age-standardised
incidence rates of renal pelvis cancer
were observed in New South Wales and
Queensland. This was particularly true in
women with ASR almost twice as high as in
the other states (Table 1). It resulted in higher
rates or renal pelvis cancer in women than
in men at the national level [M:F IRR 0.89
(95% CI 0.78,1.02)]. Over our 25-year study
period (1983–2007), renal pelvis cancer rates
decreased significantly. The decrease was
stronger in women [-52%, AAPC -3.5% (95%
CI -4.3,-2.6)] than in men [-39%, AAPC -2.1%
(95% CI -2.8,-1.4)] and in the states with the
highest rates in 1983–1987 (New South Wales
and Queensland, see Figure 1A and Table 1.
The stronger decrease in women inversed the
sex predominance with higher rates in men
than in women in 2003–2007 [IRR 1.15 (95%
CI 0.99,1.33)].
In contrast to our observations in renal pelvis
cancer incidence trends, ureteral cancer rates
remained stable and constantly higher in
men than in women between 1983 and 2007
(Figure 1B). This was observed for Australia
as a whole and in each state separately, with
the exception of a decrease in men in South
Australia (not shown).
Joinpoint analyses (supplementary file 1)
showed that the decrease in the trends
of renal pelvis cancer in women was not
constant through time. They indicated a
slight decrease in the rates from 1983 to 1994
(non-significant) followed by a statistically
significant decrease from 1995 to 2007 [APC
-6.4% (95%CI -8.0,-4.8)]. This pattern was also
observed in New South Wales with a greater
decline of the ASR from 1996 [APC -8.1%
(95% CI -11.1,-5.1)]. We did not observe any
statistically significant changes in renal pelvis
cancer trends in men over our study period.
Our age-period and age-cohort analyses
(supplementary file 2) showed a clear cohort
effect in women with decreasing trends of
renal pelvis cancer in women born in the mid-
1910s onwards. In men, there appeared to be
a weak period effect with reduction of rates
from the mid-1990s onwards.
Table 1: Age Standardized Rates (ASR) per 100,000 and Average Annual Percent Change (AAPC) for renal pelvis
cancer between 1983 and 2007 in Australia (ages 30-79).
ASR
(1983-1987)
ASR
(2003-2007)
AAPCa
[95% CI]
Men
New South Wales
Queensland
South Australia
Victoria
Western Australia
Australia (5 registries)
2.00
2.52
1.94
1.82
1.84
2.02
1.40
1.19
1.45
1.02
1.16
1.24
-1.7 [-2.6;-0.7]
-3.0 [-4.8;-1.2]
-2.0 [-4.2;0.2]
-2.4 [-3.7;-1.1]
-2.4 [-4.6;-0.1]
-2.1 [-2.8;-1.4]
Women
New South Wales
Queensland
South Australia
Victoria
Western Australia
Australia (5 registries)
3.13
3.22
1.59
1.07
1.16
2.26
1.28
1.43
0.84
0.70
0.88
1.08
-4.1 [-5.3;-2.9]
-3.3 [-4.9;-1.8]
-3.7 [-5.9;-1.5]
-2.2 [-4.0;-0.4]
-0.5 [-2.6;1.6]
-3.5 [-4.3;-2.6]
a Average annual percent change for 1983-2007.
Calendar year
Age−standardised incidence rate
(per 100,000 person−years)
1980 1985 1990 1995 2000 2005 2010
0.1 0.2 0.5 1.0 2.0
Men Women
B

Phenacetin and incidence trends of UUT cancers
2014 Online Australian and New Zealand Journal of Public Health 3
Discussion
This is the first study to examine the long-term
incidence trends of UUT cancers in
Australia after the ban on phenacetin. We
observed a marked decrease in the incidence
of renal pelvis cancer since 1983, particularly
in women, which was more pronounced
in populations reported to have had the
highest prevalence of regular use of analgesic
mixtures containing phenacetin and the
highest incidence of phenacetin-associated
renal diseases.
In 1983–1987, we observed slightly higher
incidence rates of renal pelvis cancer in
women than in men. This unusual pattern
(female predominance) for renal pelvis cancer
was also reported prior to our study period
and suggests exposure to a risk factor which
used to be more prevalent among Australian
women than among men.9 Phenacetin is the
perfect suspect to explain these high rates,
as it was specifically marketed to women in
the 1950s and 1960s, and was shown to be
responsible for the analgesic nephropathy
epidemic in Australia in the 1970s.7 The higher
incidence rates of renal pelvis cancer in 1983–
1987 (and the following stronger decreases
in these rates) observed in Queensland and
New South Wales, the states with the highest
reported phenacetin consumption and
highest analgesic nephropathy incidence
rates, further support the phenacetin
hypothesis.7,17 Taken together, these results
strongly suggest phenacetin as the risk factor
responsible for the high incidence rates of
renal pelvis cancer observed in 1983–1987. It
is also likely that the decrease in renal pelvis
cancer rates was a consequence of the 1979
ban on phenacetin which previously had a
similarly beneficial effect on the incidence of
analgesic nephropathy in Australia.1,8
Some etiological hypotheses on UUT cancers
can also be generated from our results. The
strong cohort effect observed in renal pelvis
cancer rates in women could indicate that
the withdrawal of phenacetin positively
affected first the younger cohorts (exposed
for a shorter period of time), as previously
observed in analgesic nephropathy age-specific
incidence trends.9,18 This suggests
that the increased risk of renal pelvis cancer
would only be associated with cumulative
exposure to phenacetin, as previously
hypothesised.19,20 From our trends analyses, it
also appears that the positive impact of the
phenacetin ban was only visible after 15 to 20
years. This suggests a long latency period for
the development of phenacetin-associated
renal pelvis cancers. Finally, the unexpectedly
contrasting trends of renal pelvis and
ureteral cancer incidence could indicate that
phenacetin exposure is not associated with
ureteral cancer, and therefore a different
etiology for these cancers.
These findings should be assessed in the
context of an ecological study design. In
the absence of patients’ individual data, we
could not adjust our analyses for potential
confounders such as other risk factors for
UUT cancers. However, we believe that
these did not strongly influence our results.
Because of the consistent rarity of UUT
cancers worldwide,21 even in countries with
high smoking prevalence, it is likely that
tobacco-smoking only has a limited impact
on UUT cancer trends. In Australia, smoking
prevalence consistently decreased in men
since the mid-1940s,22 while the incidence of
renal pelvis cancer increased in the 1970s.9
The only explanation for this increase that
would imply a role for smoking would be
an increase in smoking prevalence prior
to 1940 and a long latency period for the
development of UUT cancers. However,
such hypothesis would not be consistent
with the almost simultaneous decreases in
smoking prevalence and renal pelvis cancer
incidence observed in women in the 1980s
(Figure 1A).22 Finally, as smoking prevalence
has consistently been higher in men than
in women, it is unlikely to explain the sex-specific
observations in our study, particularly
the higher incidence rates in women in the
early 1980s.
This suggests that tobacco-smoking is only
weakly associated with renal pelvis cancer
and that another factor is responsible for our
findings. Apart from phenacetin, the only
other recognised risk factor for UUT cancers is
aristolochic acid. However, to our knowledge,
population exposure to aristolochic acid has
never been reported in Australia. Although
arsenic and various occupational exposures
are also suspected to be risk factors for
these cancers,20 they are unlikely to occur
more frequently in women than in men and
would therefore not explain our sex-specific
findings. Overall, and although the risk of
ecological fallacy cannot be completely
avoided, we believe that the heavy marketing
of phenacetin to women in Australia, the
availability of some data on the prevalence of
use of analgesics in the Australian population,
the sex-specific findings of our study and
the many similarities between renal pelvis
cancer and analgesic nephropathy incidence
trends provide a unique and favourable
setting in which to draw conclusions using an
ecological study design.
Although we used recent and high quality
incidence data from cancer registries covering
95% of the Australian population, such data
also have their limitations. As can be observed
with bladder cancer, changes in diagnostic
procedures or in registration practices may
create artefacts influencing incidence trends
derived from cancer registry data.9,23 However,
most of these changes would affect both
sexes similarly and would therefore not
explain our sex-specific findings. We also note
that complete national registration of cancer
was not available prior to 1983 and that the
inclusion of earlier data could have slightly
influenced our trends analyses. Finally, small
numbers of cases, particularly for ureteral
cancer, may have increased random variation
in our analyses and made it impossible to
perform more detailed analyses exploiting
age, period and cohort simultaneously.
Conclusions and implications
Our findings strongly suggest that regular
use of analgesic mixtures containing
phenacetin led to the high incidence of renal
pelvis cancer in Australia in the early 1980s,
particularly in women, and illustrate the
beneficial public health impact of the ban on
phenacetin.
Studies describing the impact of public
health prevention policies using population-level
statistics are rare and their results often
difficult to interpret as definitive conclusions
on causation cannot be drawn. In the case
of cancer studies, such assessment is further
hampered by the high number of risk
factors involved (and their interactions), by
the frequent absence of population-level
exposure data and by the long lag-time
required for the development of cancer.
However, and despite these limitations,
this work provides a case study of how a
prevention policy may successfully alleviate
a public health problem. It also offers a valid
solution to current public health issues such
as, but not limited to, the abuse of codeine-based
analgesics or of the so-called ‘legal-highs’.
24,25
Finally, this study highlights the important
role of cancer registries, not only in
monitoring cancer incidence trends and
generating etiologic hypotheses, but also
in evaluating the impact of public health
prevention measures on cancer control.

4 Australian and New Zealand Journal of Public Health 2014 Online
Acknowledgements
The authors would like to acknowledge the
directors and staff members of the cancer
registries who provided the data used in this
publication, particularly Sanchia Aranda,
Jane Walker (New South Wales), Joanne
Aitken, Carly Scott (Queensland), Ron Somers
(South Australia), Helen Farrugia (Victoria)
and Timothy Threlfall (Western Australia). We
also thank our IARC colleagues Neela Guha,
Melina Arnold and Karen Muller for their
insightful comments on the manuscript.
References
1. Mihatsch MJ, Khanlari B, Brunner FP. Obituary to
analgesic nephropathy-an autopsy study. Nephrol Dial
Transplant. 2006;21:3139-45.
2. McLaughlin JK, Lipworth L, Chow WH, Blot WJ.
Analgesic use and chronic renal failure: A critical review
of the epidemiologic literature. Kidney Int. 1998;54:
679-86.
3. Jensen OM, Knudsen JB, Tomasson H, Sorensen BL.
The Copenhagen case-control study of renal pelvis
and ureter cancer: Role of analgesics. Int J Cancer.
1989;44:965-8.
4. McCredie M, Stewart JH, Day NE. Different roles for
phenacetin and paracetamol in cancer of the kidney
and renal pelvis. Int J Cancer. 1993;53:245-9.
5. International Agency for Research on Cancer Working
Group. Overall Evaluations of Carcinogenicity: An
Updating of IARC Monographs Volumes 1 to 42. Lyon
(FRA): IARC; 1987.
6. International Agency for Research on Cancer
Working Group. A Review of Human Carcinogens:
Pharmaceuticals. Lyon (FRA): IARC; 2012.
7. Stewart JH. Analgesic abuse and renal failure in
Australasia. Kidney Int. 1978;13:72-8.
8. Michielsen P, de Schepper P. Trends of analgesic
nephropathy in two high-endemic regions with
different legislation. J Am Soc Nephrol. 2001;12:550-6.
9. McCredie M, Stewart J, Smith D, Supramaniam R,
Williams S. Observations on the effect of abolishing
analgesic abuse and reducing smoking on cancers of
the kidney and bladder in New South Wales, Australia,
1972-1995. Cancer Causes Control. 1999;10:303-11.
10. Australian Bureau of Statistics. 3101.0 - Australian
Demographic Statistics 2012. Canberra (AUST): ABS;
2012.
11. Segi M. Cancer Mortality for Selected Sites in 24 Countries
(1950-1957). Sendai (JPN): Tohoku University School of
Public Health; 1960.
12. Boyle P, Parkin DM. Statistical Methods for Registries.
In: Jensen OM, Parkin DM, MacLennan R, Muir CS,
Skeet RG, editors. Cancer Registration: Principles and
Methods. Lyon (FRA): International Agency for Research
on Cancer; 1991. p. 126-58.
13. Clegg LX, Hankey BF, Tiwari R, Feuer EJ, Edwards BK.
Estimating average annual per cent change in trend
analysis. Stat Med. 2009;28:3670-82.
14. Joinpoint Regression Program. Version 4.0.4. Bethesda
(MD): National Cancer Institute Statistical Methodology
and Applications Branch, Surveillance Research
Program; 2013.
15. Clayton D, Schifflers E. Models for temporal variation
in cancer rates. I: Age-period and age-cohort models.
Stat Med. 1987;6:449-67.
16. Loader C. Smoothing: Local Regression Techniques 2004.
Berlin (DEU): Humboldt-University of Berlin Centre for
Applied Statistics and Economics; 2004 [cited 2013 Oct
10]. Available from: http://hdl.handle.net/10419/22186
17. McCredie M, Coates MS, Ford JM, Disney AP, Auld JJ,
Stewart JH. Geographical distribution of cancers of the
kidney and urinary tract and analgesic nephropathy in
Australia and New Zealand. Aust N Z J Med. 1990;20:684-
8, 694.
18. Chang SH, Mathew TH, McDonald SP. Analgesic
nephropathy and renal replacement therapy in
Australia: Trends, comorbidities and outcomes. Clin J
Am Soc Nephrol. 2008;3:768-76.
19. McCredie M, Ford JM, Taylor JS, Stewart JH. Analgesics
and cancer of the renal pelvis in New South Wales.
Cancer. 1982;49:2617-25.
20. Colin P, Koenig P, Ouzzane A, Berthon N, Villers A,
Biserte J, et al. Environmental factors involved in
carcinogenesis of urothelial cell carcinomas of the
upper urinary tract. BJU Int. 2009;104:1436-40.
21. Ferlay J, Parkin DM, Curado MP, Bray F, Edwards B, Shin
HR, et al. Cancer Incidence in Five Continents, Volumes I
to IX: IARC Cancer Base No. 9. Lyon (FRA): International
Agency for Research on Cancer; 2010 [cited 2013 Oct
10]. Available from: http://ci5.iarc.fr
22. Scollo MM, Winstanley MH. Tobacco in Australia: Facts
and Issues. 4th ed. Melbourne (AUST): Cancer Council
Victoria; 2012 [cited 2013 Oct 10]. Available from: http://
www.TobaccoInAustralia.org.au/
23. Luke C, Tracey E, Stapleton A, Roder D. Exploring
contrary trends in bladder cancer incidence, mortality
and survival: Implications for research and cancer
control. Intern Med J. 2010;40:357-62.
24. McAvoy BR, Dobbin MD, Tobin CL. Over-the-counter
codeine analgesic misuse and harm: Characteristics
of cases in Australia and New Zealand. N Z Med J.
2011;124:29-33.
25. Handley SA, Flanagan RJ. Drugs and other chemicals
involved in fatal poisoning in England and Wales during
2000-2011. Clin Toxicol (Phila). 2014;52:1-12.
Supporting Information
Additional supporting information may be
found in the online version of this article:
Supplementary Figure 1: Joinpoint analyses
for renal pelvis cancer rates in women aged
30-79 (1983-2007) in Australia and New South
Wales.
Supplementary Figure 2: Age-Period and
Age-Cohort analyses of the incidence of renal
pelvis cancer in men and women in Australia
between 1983 and 2007 in 5-year age groups
(only shown for 55-79 years old).
Antoni et al.

The ban on phenacetin is associated with changes in the incidence trends of upper-urinary tract cancers in Australia

More Related Content

What's hot (20)

Viewers also liked (15)

Similar to The ban on phenacetin is associated with changes in the incidence trends of upper-urinary tract cancers in Australia (20)

More from Cancer Council NSW (8)

Recently uploaded (20)

The ban on phenacetin is associated with changes in the incidence trends of upper-urinary tract cancers in Australia