FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`(2009) 20:935-944
`Cancer Causes Control
`DOI 10.1007/s10552-009-9328-9
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`
`Pleural
`and
`
`and
`
`peritoneal
`
`mesotheliomas
`
`in
`
`SEER:
`
`age
`
`effects
`
`temporal
`
`trends,
`
`1973-2005
`
`H. Moolgavkar
`Suresh
`Jay Turim
`
`" Rafael Meza
`
`"
`
`26 February 2009/Published
`Received: 12 October 2008/Accepted:
`© Springer Science+Business Media B.V. 2009
`
`online: 18 March 2009
`
`We analyzed
`in the
`incidence
`mesothelioma
`Abstract
`Results
`End
`and
`Epidemiology,
`(SEER)
`Surveillance,
`of
`the
`extensions
`1973-2005
`the period
`over
`database
`using
`the
`In
`these
`models.
`age-period-cohort
`analyses,
`(APC)
`conventional
`APC
`the
`age effects
`of
`usual
`non-specific
`from
`functions
`derived
`were
`replaced
`by hazard
`models
`the Armitage-
`models
`of
`two multistage
`carcinogenesis,
`clonal
`expansion
`Doll model
`and the two-stage
`(TSCE)
`APC models
`described
`the incidence
`model.
`The extended
`After
`and peritoneal
`mesotheliomas
`well.
`data on pleural
`suggest
`the data
`that
`the
`adjustment
`for
`temporal
`trends,
`rates of both
`pleural
`and peritoneal
`age-specific
`incidence
`in men
`and women.
`Driven
`are identical
`mesotheliomas
`age-adjusted
`rates of pleural
`by birth
`cohort
`effects,
`largely
`7.5 per mil-
`from about
`mesothelioma
`among men
`rose
`person-
`20 per million
`in 1973 to about
`lion
`person-years
`1990s and appear
`to be stable or declining
`years in the early
`of
`pleural
`mesothelioma
`thereafter.
`Age-adjusted
`rates
`at
`remained
`more
`or
`less
`constant
`women
`have
`among
`1973-
`the period
`2.5 per million
`person-years
`over
`about
`in
`mesothelioma
`Age-adjusted
`rates
`for
`peritoneal
`2005.
`women
`men
`(1.2 per million
`and
`both
`person-years)
`no temporal
`trends
`(0.8 per million
`exhibit
`person-years)
`
`this
`The online version of
`material
`Electronic
`supplementary
`contains supplementary
`article (doi:10.1007/s10552-009-9328-9)
`material, which is available to authorized users.
`
`S. H. Moolgavkar
`inc., Benevue, WA, USA
`Exponent
`
`. R. Meza · J. Turim
`S. H. Moolgavkar
`(B)
`Fred Hutchinsons Cancer Research Center, Seattle, WA, USA
`e-mail: smoolgav@fherc.org
`
`J. Turim
`Exponent
`
`Inc., Alexandria, VA, USA
`
`over
`
`of
`the
`the period
`cases
`94,000
`mately
`mesothelioma
`peritoneal
`period
`2005-2050.
`
`study. We estimate
`that
`cases
`of
`pleural
`and
`15,000
`will
`occur
`in the US over
`
`approxi-
`
`of
`the
`
`Keywords
`Age-specific
`
`Mesothelioma
`incidence
`
`. Asbestosis
`· Multistage
`carcinogenesis
`
`.
`
`Introduction
`
`Mesothelioma
`for
`
`the
`
`called
`Multiple
`
`sentinel malig-
`been
`has often
`epidemiological
`asbestos
`exposure.
`nancy
`exposed
`cohorts
`of workers
`studies
`occupationally
`among
`of mesothelioma.
`reported
`elevated
`risks
`to asbestos
`have
`pure chrys-
`While
`there
`is still
`as to whether
`controversy
`can cause
`otile exposure,
`uncontaminated
`with
`amphiboles,
`mesothelioma
`the
`evidence
`shows
`that
`2],
`[1,
`clearly
`are far more
`potent
`than chrysotile
`in causing
`amphiboles
`have
`mesothelioma
`[3-5].
`Case-control
`studies
`[6,
`7]
`expo-
`certain
`shown
`that
`occupations
`involving
`similarly
`risk
`sure to asbestos
`are associated
`with
`an increased
`the fiber
`mesothelioma.
`type, whether
`chrysotile,
`However,
`of
`or
`a mixture
`the
`is not
`amphiboles,
`two,
`generally
`reported
`in these studies.
`.
`trends m
`the temporal
`Price
`and Ware
`showed
`[8]
`temporal
`males
`reflect
`mesothelioma
`incidence
`among
`a lag of
`in asbestos
`use in the United
`trends
`States with
`vears.
`used
`an age-cohort
`model
`to analyze
`20-40
`They
`incidence
`in the Surveillance,
`mesothelioma
`Epidemiology,
`the
`period
`and
`End
`Results
`database
`over
`(SEER)
`Their
`age-adjusted
`1973-2000.
`results
`suggest
`that
`rates
`reached
`a peak around
`the early
`to mid-1990s
`among males
`some-
`and have
`remained
`constant
`or declined
`relatively
`reported
`the age-adjusted
`what
`since
`then. They
`that
`
`that
`
`of
`
`rates
`
`Springer
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`936
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`Cancer Causes Control
`(2009) 20:935-944
`
`We have previously
`analyses
`likelihood-based
`presented
`of different
`the incidence
`cancers
`in the SEER registry
`to
`hazard
`functions
`derived
`from multistage
`models
`using
`the non-specific
`of age in the traditional
`replace
`effects
`while
`secular
`and cohort
`APC models,
`i.e., period
`trends,
`were modeled
`in the usual
`fashion
`(see [12-14]).
`effects,
`we model
`the
`(age
`age-specific
`incidence
`Briefly,
`in calendar
`year
`f as
`occurring
`h, (a) = b,c h(a),
`from a multistage
`function
`is the hazard
`where
`derived
`h(a)
`and
`that adjusts
`for calendar
`cjis
`a coefficient
`yearl,
`model,
`(i = f
`for
`birth
`the
`coefficient
`b, adjusts
`cohort
`i
`a,
`1890-1894,...,1955-
`stratifiedin5-yeargroups;
`1885-1889,
`1990, and >1960).
`to the SEER database
`Conforming
`mat, we
`the
`data
`in
`18 age groups
`(0-4
`years,
`stratify
`seven peri-
`5-9
`years,...,80-84
`and into
`85+
`years,
`years)
`then
`ods
`1976-1980,...,2001-2005).
`We
`fit
`(1973-1975,
`each of
`three multistage
`of observed
`models
`to the number
`and calendar
`mesothelioma
`cases
`stratified
`by age group
`year. We obtained
`parameter
`estimates
`for each model
`by
`the
`likelihood
`across
`all
`age-calendar
`strata
`maximizing
`or perito-
`the number
`of cases, either
`that
`pleural
`assuming
`in each
`stratum
`is Poisson
`distributed
`with mean
`neal,
`* hq(a), where Ng is the population
`at risk
`in age group
`Ng
`i and birth
`is as defined
`and hq(a)
`above.
`cohort),
`We considered
`the
`three
`distinct
`multistage
`models
`age effects.
`we considered
`the Armitage-Doll
`First,
`[16]
`function
`model
`of carcinogenesis.
`In this model,
`the hazard
`for mesothelioma
`is a power
`considered
`of age. Peto [17]
`similar
`model
`for mesothelioma
`in occupational
`cohorts
`exposed
`in which
`to asbestos,
`the
`hazard
`function
`is a
`exposure. We discuss
`of
`power
`since
`this model
`time
`first
`below. We also considered
`the two-stage
`expansion
`clonal
`proposed
`(TSCE) model
`by Moolgaykar
`and Venzon
`[18]
`and Moolgavkar
`and
`Knudson
`and
`a three-stage
`[19]
`extension
`of
`it as described
`in earlier
`[12-14].
`papers
`
`a)
`
`-
`
`for-
`
`for
`
`a
`
`of
`
`and have remained more or less
`are much lower
`females
`for
`1973-2000.
`In a
`of
`the analyses,
`constant
`over
`the period
`[9] examined
`the SEER data
`more
`recent
`paper, Teta et al.
`five-year
`over
`age groups
`the period
`1973-2002
`and
`by
`the temporal
`trends of
`incidence
`rates among
`that
`concluded
`older
`than 60.
`men were driven
`by age groups
`largely
`is
`The most
`common
`site for malignant
`mesothelioma
`of
`but
`significant
`a small
`fraction
`the pleura.
`However,
`In this paper, we extended
`cases occurs
`in the peritoneum.
`and Ware
`in two ways.
`First, we
`the analyses
`of Price
`[8]
`sepa-
`mesotheliomas
`considered
`pleural
`and
`peritoneal
`age-period-cohort
`rately.
`we
`conducted
`a full
`Second,
`that
`It
`for each of
`the sites.
`is well
`known
`analysis
`(APC)
`are used in
`all
`three factors,
`and cohort,
`when
`age, period,
`problem of
`there is a fundamental
`of
`analyses
`data,
`registry
`sometimes
`referred
`to as the
`[10,
`11],
`non-identifiability
`in the data. We finessed
`problem
`of arbitrary
`linear
`trend
`that
`age is the fundamental
`this
`problem
`by
`recognizing
`and period
`determinant
`of cancer
`incidence
`and that cohort
`the effect
`we replaced
`effects modulate
`of age. Specifically,
`the non-specific
`age effects
`of
`traditional
`APC models
`by
`of
`parametric
`functions
`derived
`from multistage
`models
`We have previously
`for
`used this approach
`carcinogenesis.
`13], pan-
`of
`the incidence
`of colon
`cancer
`analyses
`[12,
`and esophageal
`cancer
`in SEER.
`creatic
`cancer
`[13],
`[14]
`
`Materials
`
`and methods
`
`and peri-
`the pleura
`of
`data for mesotheliomas
`Incidence
`for
`the
`obtained
`from the SEER
`toneum
`were
`9 registry
`for
`the
`1973-1992
`and from the SEER
`years
`13 registry
`Review
`See the SEER Cancer
`Statistics
`1993-2005.
`years
`for details
`this database.
`For our analyses,
`[15]
`concerning
`we used the reported
`incidence
`of mesothelioma
`by gender,
`year
`age, and calendar
`in the SEER geographic
`areas.
`from the SEER
`The
`population
`bases were
`obtained
`population
`files
`(based
`on the data from the US Census
`five-year
`sex
`and were
`cross-tabulated
`Bureau)
`by
`by
`over
`periods
`the calendar
`years
`1973-2005
`and 5-year
`age
`Model
`for
`each
`gender
`(ages
`groups
`from 0-85+).
`fits
`of 6,017
`all
`included
`races. For all years
`a total
`combined,
`and 1,673 female
`male cases (5,562
`455 peritoneal)
`pleural,
`available
`for
`the
`cases (1,291
`382 peritoneal)
`were
`pleural,
`analyses.1
`
`° Incidence data for the same tumor can be downloaded using different
`coding schemes from the SEER registry. These distinct methods of
`downloading the data should yield identical numbers. For example, one
`can download mesothelioma incidence data using either ICD 10 codes
`(C45.0 for mesothelioma of the pleura and C45.1 for mesothelioma of
`omentum and mesentery), or using site &
`the peritoneum,
`including
`omentum and mesentery'
`and histol-
`= 'peritoneum,
`morphology
`ogy = 9,050-9,059
`these
`(mesothelial
`neoplasms). For peritoneum,
`
`Springer
`
`Footnote 1 continued
`results, with the latter coding
`two codings yield highly discrepant
`times as many tumors as the former. For
`four
`yielding approximately
`to these two approaches yield
`the appropriate
`equivalents
`pleura,
`identical
`In this manuscript, we present
`numbers.
`the results of
`analyses of peritoneal mesothelioma
`incidence based on the second
`this is the older method of
`approach for
`the following
`reasons. First,
`less susceptible to error. Second,
`it
`coding and,
`therefore, probably
`appears to have been the method used in the papers by Teta et al.
`[9]
`and Boffetta [31]. We have also performed the analyses using the data
`obtained using the first method. Our conclusions regarding trends and
`fits remain unchanged. However, our estimated rates for per-
`model
`itoneal
`mesothelioma
`and
`projected
`number
`of
`cases
`are
`a fourth of those presented here. We would be happy to
`approximately
`share these results with any interested reader.
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`Cancer Causes Control
`
`(2009) 20:935-944
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`937
`
`The Armitage-Doll
`
`model
`
`(16)
`
`has the following
`This model
`function:
`specific
`incidence)
`=Ctk
`
`H(t)
`
`form for
`
`the hazard
`
`(age-
`
`where C and k are constants
`
`to be estimated
`
`and t
`
`is age.
`
`The TSCE model
`
`(18,
`
`19)
`
`The two-stage model
`via a Poisson
`initiated
`that cells
`posits
`conver-
`clonal
`and malignant
`expansion
`process
`undergo
`sion via a birth-death-mutation
`this
`of
`process.
`The details
`model
`in a number
`(18-21).
`are presented
`of publications
`for
`more
`The hazard
`function
`this model
`is considerably
`that
`than
`of
`the Armitage-Doll
`The
`complicated
`model.
`model
`has four parameters,
`the rate of
`the rate
`initiation,
`y,
`of division,
`±, and death,
`initiated
`², of
`and the rate of
`cells,
`all
`four
`can be
`malignant
`g. Not
`parameters
`conversion,
`data alone. We estimated
`estimated
`from incidence
`three
`below. With
`identifiable
`parameters
`as described
`constant
`the hazard
`function
`for
`this model
`takes
`the
`parameters,
`form:
`following
`e’-e-Pf
`
`pq
`
`,
`
`v
`
`h(t)=
`
`with
`of a quadratic
`p and q are the roots
`where
`equation,
`p+q=g=
`-(±-²-
`We estimatedp,
`µ) andpq=ag.
`q, and r s j, which
`param-
`a set of
`identifiable
`comprise
`the TSCE model
`eters. Thus,
`the estimation
`of one
`requires
`more parameter
`than the Armitage-Doll
`model. Note that g
`is roughly
`rate
`of proliferation
`of
`the net
`initiated
`cells
`(since µ is a mutation
`rate and much
`smaller
`than ± and ²),
`q ~ µ/(1-
`and r is what Fisher
`has called
`the index
`²/±),
`of diversity
`[22]. More
`are provided
`in the papers
`details
`referenced
`above
`19-21].
`[12-14,
`
`to
`
`of
`
`We also fit a three-stage
`the TSCE model
`extension
`the data as described
`in earlier
`[12-14].
`papers
`We fit models
`In
`to the data by maximizing
`likelihoods.
`the current
`context, models
`from the use of distinct
`arising
`We
`multistage
`hazard
`functions
`are
`not
`hierarchical.
`therefore
`used the Akaike
`Criterion
`to
`Information
`(AIC)
`judge
`the relative
`fits of
`the different
`models.
`All
`and
`analyses
`were
`conducted
`for pleural
`separately
`peritoneal
`mesotheliomas.
`At each site, we first
`fit models
`data. We began by fitting
`to male and female
`the
`separately
`APC model
`conventional
`in which
`separate
`parameters
`were
`fit
`for each age group,
`(not
`birth
`and period
`cohort,
`non-
`shown).
`Although
`this model
`from parameter
`suffers
`and
`the expectation
`is well
`defined
`problems,
`identifiability
`likelihood
`the maximized
`is unique.
`As judged
`by the AIC,
`in which
`non-specific
`models
`age parameters
`were replaced
`by hazard
`functions
`from multistage
`than
`did better
`models
`the conventional
`APC model
`(see Table
`1). Moreover,
`the
`three-stage
`extension
`of
`the TSCE model
`than
`did no better
`the TSCE model
`and we therefore
`do not consider
`it
`further.
`The TSCE model
`described
`the data better
`than
`consistently
`the Armitage-Doll
`model
`as judged
`and we
`the AIC
`by
`focus here on results
`of analyses
`based on the TSCE model.
`model
`since the Armitage-Doll
`of
`has the virtue
`However,
`and simplicity
`has been in use
`and, moreover,
`transparency
`for more
`than
`two
`discuss
`decades
`we
`the
`age-specific
`incidence
`estimated
`from that model
`as well.
`
`Mesothelioma
`
`projections
`
`We projected
`to the year 2050 in
`incidence
`mesothelioma
`the United
`similar
`States
`using methods
`to those in Price
`the US year 2002 pop-
`and Ware
`[8]. First, we projected
`ulation
`the future
`into
`rates
`from the US
`using mortality
`sta-
`for
`decennial
`tables
`1989-1991
`life
`[23]. New births
`tistics
`from 2002
`to 2005 were
`obtained
`from the CDC
`
`Table 1 Parameter estimates and 95% confidence intervals of the TSCE and Armitage-Doll
`incidence in SEER 1973-2005
`
`models fits to pleural and peritoneal mesothelioma
`
`Model
`
`TSCE
`
`Armitage-Doll
`
`Parameter
`
`r
`-p
`
`AIC2
`
`C
`k
`AIC'
`
`Pleura
`2.80 x 10"
`(2.30, 3.41) × 10-4
`0.12 (0.114, 0.127)
`1.40 x 10-5 (1.04, 1.90) x 10-5
`-24603
`3.36 × 10-15 (1.41, 7.99) × 10-15
`5.14 (4.95, 5.34)
`-24532.2
`
`Peritoneal
`3.17 × 10-5
`(2.41, 4.18) × 10-5
`0.11 (0.096, 0.13)
`1.78 × 10-4
`(0.98, 3.24) × 10-4
`1390.6
`(0.56, 5.50) × 10-11
`1.75 × 10-11
`2.79 (2.53, 3.08)
`1424.4
`
`Wald confidence intervals were estimated on logit
`* Akaike information
`criterion
`peritoneal mesothelioma = 1429.7
`= -24578.24,
`Age-cohort model AIC: pleural mesothelioma
`peritoneal mesothelioma = 1446.4
`= -24576.74,
`model AIC: pleural mesothelioma
`Age-period-cohort
`
`transformed parameters and then back transformed to the original scale
`
`Springer
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`938
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`Cancer Causes Control
`(2009) 20:935-944
`
`Vital
`National
`n chs/births.htm
`that
`the number
`2005.
`Future
`assumed
`
`Statistics
`and
`[24]).
`of births
`
`System (see http://www.cde.gov/
`For
`years, we assumed
`future
`is equal
`to those in
`in each year
`
`-
`
`-
`
`-
`
`Pleura
`TSCE
`AD
`
`d -
`
`s
`
`-
`
`-
`
`-
`
`Peritoneum
`TSCE
`AD
`
`and calendar
`cohort
`birth
`the last estimated
`to equal
`
`coefficients
`
`were
`
`O
`
`year
`value.
`
`Results
`
`the
`with
`together
`estimates
`the parameter
`1 shows
`Table
`and TSCE models. We also fit
`the Armitage-Doll
`AIC of
`to the male and
`and APC models
`age-cohort
`the traditional
`these
`and computed
`the AICs
`for
`female
`data sets separately
`models.
`The AICs
`for
`the conventional
`age-cohort
`and APC
`are reported
`in the legend
`to Table
`1. Recall
`that
`the
`models
`smaller
`the AIC,
`the better
`the fit of
`the model.
`this
`By
`is the best model
`for both pleural
`the TSCE model
`criterion,
`and peritoneal
`mesotheliomas.
`The Armitage-Doll
`model
`for
`fit
`is worse
`than
`the
`conventional
`models
`pleural
`for peritoneal
`mesothelioma.
`mesothelioma
`but better
`(to the 2000 US pop-
`the age-adjusted
`Figure
`1 shows
`in SEER over
`incidence
`the period
`ulation) mesothelioma
`1973-2005
`and
`the
`predictions
`made
`our
`preferred
`(see below).
`models
`
`by
`
`Pleural mesothelioma
`
`of pleural mesothelioma
`the incidence
`on the AIC,
`Based
`postulates
`best
`in SEER is described
`by a model
`that
`
`a
`
`e -
`
`women
`
`o
`
`°
`
`a
`
`Pleura
`
`u
`
`o
`
`a
`
`e -
`
`1975
`
`1980
`
`1985
`
`1990
`
`1995
`
`2000
`
`2005
`
`Peritoneum
`
`°
`
`a
`
`o o
`
`o
`
`o
`
`o
`
`8 °
`
`°
`
`o
`
`o
`
`o
`
`a
`
`0
`
`o
`
`o o
`o o o o
`
`n
`
`°
`
`o
`
`o
`
`o
`
`1975
`
`1980
`
`1985
`
`1990
`Year
`
`1995
`
`2000
`
`2005
`
`.
`.
`Fig. 1 Observed and expected age-adjusted mcidence rates over the
`lower panel:
`period 1973-2005. Upper panel: pleural mesothelioma;
`peritoneal mesothelioma
`
`Springer
`
`°
`
`-
`
`2 y -
`
`o
`
`'
`
`,
`
`- '
`
`a
`
`0
`
`20
`
`40
`Age
`
`80
`
`so
`
`o
`
`20
`
`so
`
`so
`
`40
`Age
`
`.
`.
`Age-specific mcidence curves generated by the Armitage-
`Fig. 2
`Doll and TSCE models. These curves represent
`the estimated hazard
`functions of
`these models multiplied
`by 100,000. Left panel: pleural
`mesothelioma. See the text
`for details of
`the constraints required for
`of
`the incidence curve for pleural mesothelioma. Right
`identifiability
`panel: peritoneal mesothelioma. No constraints are required for
`the
`generation of
`these curves since cohort effects for women are all
`equal
`to 1
`
`2
`
`and
`curve
`incidence
`age-specific
`common
`for males
`birth
`separate
`females, with
`in the
`and period
`effects
`cohort
`two
`sexes.
`Figure
`1 shows
`the
`in the
`age-adjusted
`rates
`SEER data together
`with
`the rates predicted
`by the TSCE
`model.
`This
`figure
`shows
`that
`incidence
`the age-adjusted
`among males
`rose rapidly
`from 1973 to about
`1990 when it
`a peak. There
`reached
`is a hint of a decline
`in the
`beginning
`1990s.
`The
`age-adjusted
`incidence
`females
`early
`among
`to be more or
`appears
`less constant
`over
`the entire period
`of
`Figure
`observation.
`shows
`the
`estimated
`age-specific
`Tem-
`mcidence
`curves
`the TSCE
`and Peto models.
`using
`poral
`trends
`are
`dominated
`cohort
`effects,
`by
`strong
`3).
`men
`(Fig.
`The
`cohort
`first
`three
`among
`eSPecially
`in
`females
`effects
`estimates
`to
`one
`were
`almost
`equal
`another
`and had large
`confidence
`as did the last
`intervals,
`three.
`In the final models,
`the first
`three cohort
`effects
`for
`females were set equal
`to one another
`as were the last
`three,
`to a total of 12 cohort
`effects
`parameters
`in females
`leading
`and
`16 in males.
`The
`cohort
`effects
`are modulated
`by
`(Fig.
`or
`period
`effects
`that
`are more
`less constant
`for
`4)
`maleS, but decline modestly
`for
`females
`(Fig.
`5). Figure
`6
`shows
`the lifetime
`of developing
`mesothelioma
`probability
`by birth
`year
`adjusted
`for
`survival
`of
`from other
`causes
`.
`Fig-
`(US decenmal
`life
`tables
`for
`1989-1991).
`mortality
`ure 7 shows
`the lifetime
`unadjusted
`for other
`
`probability
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`Cancer Causes Control
`
`(2009) 20:935-944
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`939
`
`indicate
`
`Among
`lifetime
`
`birth
`which
`
`and female
`the earliest
`
`lifetime
`and latest
`
`probabilities
`birth
`cohorts.
`
`are virtually
`
`identical
`
`in
`
`Peritoneal
`
`mesothelioma
`
`incidence
`The
`is described
`
`of
`
`peritoneal
`a model
`
`by
`
`best
`
`mesothelioma
`that
`postulates
`
`SEER
`in
`common
`
`that, over
`the period
`meso-
`of pleural
`increase
`
`by
`
`These figures
`cause mortality.
`the lifetime
`of observation,
`probability
`a rapid
`showed
`thelioma
`among men
`until
`the cohort
`of
`the early
`cohort
`1920s,
`following
`there was an equally
`rapid
`decline.
`These
`trends
`clearly
`reflect
`in birth
`cohort
`effects
`shown
`in Fig.
`3.
`the trends
`is
`there
`a more modest
`increase
`and
`females,
`risks appear
`to have stabilized.
`male
`Interestingly,
`
`Pleura
`
`- Women
`
`Peritoneum
`
`- Men
`
`1910
`
`1920
`
`1930
`
`1940
`
`1950
`
`1960
`
`Pleura
`
`- Men
`
`i
`
`1900
`
`1920
`Birth Year
`
`1940
`
`1960
`
`Fig. 3 Birth--cohort effects and 95% confidence intervals for pleural
`lower panel: men
`mesothelioma. Upper panel: women;
`
`Pleura
`
`- Women
`
`1900
`
`1910
`
`1920
`
`1940
`
`1950
`
`1960
`
`1930
`Birth Year
`
`Fig. 5
`Birth-Ä:ohort
`peritoneal mesothelioma
`
`and 95% confidence
`effects
`among men
`
`intervals
`
`for
`
`Pleural Mesothelioma
`
`---
`
`men
`
`1975
`
`1980
`
`1985
`
`1990
`
`1995
`
`2000
`
`-
`
`. .
`
`Pleura
`
`- Men
`
`à-
`m
`E
`
`1900
`
`1920
`Birth Year
`
`1940
`
`1960
`
`Peritoneal
`
`Mesothelioma
`
`1975
`
`1980
`
`1990
`1985
`Calendar Year
`
`1995
`
`2000
`
`intervals
`Period effects and 95% confidence
`Fig. 4
`lower panel: men
`mesothelioma. Upper panel: women;
`
`for pleural
`
`,--
`
`_ _ _ _ _
`
`-----------
`
`--
`
`- -
`
`- - - -
`
`.
`
`i
`1900
`
`I
`1910
`
`I
`1920
`
`I
`1940
`
`I
`1950
`
`I
`1960
`
`I
`1930
`Birth Year
`
`Fig. 6
`of mesothelioma
`Lifetime
`by birth cohort after
`probability
`for other cause mortality. Upper panel: pleural mesothe-
`adjustment
`lower panel: peritoneal mesothelioma
`lioma;
`
`Springer
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`940
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`(2009) 20:935-944
`Cancer Causes Control
`
`o -
`-
`
`Pleural Mesothelioma
`
`-
`Men
`- - Women
`
`8 -
`
`Projected
`
`Number
`
`of Mesothelioma
`
`- US
`Cases
`--- h
`- - Women
`
`<
`-o
`
`1900
`
`1920
`Birth Year
`
`1940
`
`1960
`
`Peritoneal
`
`Mesothelioma
`
`2010
`
`2020
`
`2030
`Year
`
`2040
`
`2050
`
`1900
`
`1910
`
`1920
`
`1930
`Birth Year
`
`1940
`
`1950
`
`1960
`
`2010
`
`2020
`
`2030
`Year
`
`2040
`
`2050
`
`of mesothelioma by birth cohort without
`Lifetime probability
`Fig. 7
`for other cause mortality. Upper panel: pleural mesothe-
`adjustment
`lower panel: peritoneal mesothelioma
`lioma;
`
`Fig. 8 Projected number of mesothelioma cases in the US to the year
`2050. Upper panel: pleural mesothelioma;
`lower panel: peritoneal
`mesothelioma
`
`incidence
`
`rates in males and females with birth
`age-specific
`equal
`females
`to one among
`and
`effects
`cohort
`identically
`equal
`to one in both genders.
`effects
`The
`period
`identically
`confi-
`have wide
`among males
`estimated
`cohort
`effects
`similar
`to the cohort
`but suggest
`a pattern
`dence intervals,
`for pleural mesothelioma
`among males.
`the
`effects
`Thus,
`in peritoneal
`with
`no secular
`trends
`data
`are consistent
`incidence
`women
`and
`mesothelioma
`weak
`only
`among
`over
`the period
`1973-2005.
`trends
`Figure
`among men
`curves
`age-specific
`incidence
`for
`shows
`the
`peritoneal
`generated
`the Armitage-Doll
`and
`mesothelioma
`the
`by
`TSCE models.
`there
`have been no temporal
`Since
`trends,
`the lifetime
`of peritoneal
`mesothelioma
`probability
`among
`for other
`is constant
`adjusted
`cause mortality,
`at
`women,
`1 × 10-4
`over
`the birth
`cohorts
`in the
`all
`approximately
`1 × 10-4
`lies between
`data. Among men,
`this probability
`and 1.5 × 10-4
`for all birth
`cohorts
`(Fig.
`6). The lifetime
`for other
`cause mortality
`(to age 85) unadjusted
`probability
`1 × 10-4
`women
`and
`ranges
`is about
`between
`among
`1 × 10-4
`and 2.5 x 10-4
`among men (Fig.
`7).
`
`2
`
`and perito-
`cases of pleural
`incident
`of
`to the year 2050 are
`in the US projected
`
`Projected
`
`incidence
`
`The total
`numbers
`neal mesothelioma
`shown
`in Fig.
`8.
`
`Springer
`
`Discussion
`
`In this paper, we have used APC models
`age
`to investigate
`and temporal
`in mesothelioma
`in
`effects
`trends
`incidence
`SEER, with
`the non-specific
`age effects
`APC
`of
`traditional
`models
`replaced
`by parametric
`incidence
`curves
`based on
`the Armitage-
`the ideas of multistage
`carcinogenesis.
`Both
`and TSCE models
`Doll
`describe
`the data well within
`this
`by this pro-
`analytic
`framework.
`The
`age effects
`isolated
`approximate
`the
`age-specific
`incidence
`curves
`of
`cedure
`pleural
`and peritoneal
`mesotheliomas
`adjustment
`for
`after
`temporal
`trends
`that are largely
`due to asbestos
`exposure,
`be involved. We have
`although
`other
`factors
`could
`also
`used
`a number
`of
`assumptions
`mesothelioma
`to project
`incidence
`in the US to the year 2050.
`
`Age-specific
`
`incidence
`
`by the two
`As seen in Fig.
`predicted
`the incidence
`curves
`2,
`pre-
`model
`with
`the Armitage-Doll
`are similar,
`models
`the older
`ages.
`somewhat
`higher
`rates at
`incidence
`dicting
`the Armitage-
`For pleural mesothelioma,
`the exponent
`k of
`in Table
`Doll model
`is estimated
`1.
`to be ~5
`as shown
`Peto et al.
`in an analysis
`the insulator
`database
`[17]
`found
`that
`the mortality
`rates
`those
`continuously
`among
`exposed was described
`well
`by a model
`in which
`duration
`of exposure
`is lagged
`by 10 years
`and raised
`to the power
`3.2. However,
`in a document
`for
`prepared
`the U.S. EPA in
`1986
`Nicholson
`pointed
`out
`et al.
`that Peto
`[25],
`[16]
`
`of
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`Cancer Causes Control
`
`(2009) 20:935-944
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`941
`
`1922
`before
`the workforce
`entered
`who
`workers
`excluded
`the
`over
`the age of 80. For
`1946 and who were
`and after
`that a model with
`time since
`found
`entire cohort, Nicholson
`the data well.
`raised
`first exposure
`to the power
`5 described
`of mesot-
`estimated
`the number
`Nicholson
`et al.
`[26]
`that
`could
`be expected
`for different
`occupations
`heliomas
`from 1980
`through
`data
`from the
`insulator
`2030,
`using
`cohort
`the parameters
`of
`the mortality
`rate as
`to estimate
`function
`of
`time
`since
`first
`which
`he
`also
`exposure,
`assumed
`to be proportional
`to time
`raised
`to a power.
`He
`found
`that
`of death
`the risk
`increased
`as the fourth
`or
`fifth
`or
`power
`of
`time
`from onset
`of exposure
`for
`about
`40
`50 years, without
`a precise
`value
`of
`the exponent.
`giving
`We estimate
`from the
`data
`presented
`in the Nicholson
`the exponent
`is about
`4.5. Thus,
`the value
`of
`the
`paper
`that
`by the modi-
`the Armitage-Doll
`model
`given
`exponent
`of
`here
`fied APC
`approach
`used
`is
`consistent
`with
`values
`found
`in the literature.
`It should
`be kept
`in mind,
`however,
`that
`in the model
`here
`it
`is the age that
`is raised
`to the
`power
`in the analyses
`of Peto and Nicholson
`it
`5, whereas
`is time
`since first exposure.
`cancer. Doll
`with
`obtains
`An
`situation
`analogous
`lung
`rates
`that
`lung cancer mortality
`and Peto [27] have reported
`with
`between
`the fourth
`and fifth
`power
`of age
`increase
`and with
`a similar
`power
`of duration
`non-smokers
`among
`of smoking
`smokers.
`The multiplicative
`parameter
`among
`C in the Armitage-Doll
`model
`is not uniquely
`identifiable
`because
`in an APC model
`the estimate
`of
`this
`parameter
`is
`of
`the cohort
`or period
`parameters
`depends
`upon which
`although
`v/± is an
`anchored
`at 1. For
`the
`same
`reason,
`the TSCE model, when
`the TSCE
`identifiable
`parameter
`of
`model
`is embedded
`in an APC model,
`this
`parameter
`is
`identifiable
`up to a multiplicative
`non-zero
`constant.
`only
`1926-
`for
`cohort
`In these analyses,
`the parameters
`the birth
`1930 for women
`and the period
`2001-2005
`for men were
`constrained
`to be 1. Then
`the parameters
`C and v/± are
`in Fig.
`2 are
`The age-specific
`incidence
`curves
`identifiable.
`based on this choice
`of constraints.
`to 3
`is close
`For peritoneal
`mesothelioma
`the exponent
`for pleural mesotheli-
`and so is smaller
`than the exponent
`There
`are no analyses
`of peritoneal
`mesothelioma
`oma.
`dis-
`comparable
`to the
`analyses
`of Peto
`and Nicholson
`cussed
`above.
`the Armitage-Doll
`framework
`Within
`the
`model,
`and
`four
`six
`that
`suggest
`these
`results
`approximately
`of
`in
`genesis
`the
`involved
`are
`mutations,
`respectively,
`the con-
`mesotheliomas.
`pleural
`and peritoneal
`Moreover,
`the number
`of cells at risk
`stant C, which
`is the product
`of
`larger
`in peritoneal
`and the mutation
`is considerably
`rates,
`than in pleural mesothelioma
`(Table
`1). This
`is to
`finding
`the
`product
`six mutation
`be expected
`since C involves
`than
`1, for pleural mesothelioma,
`rates, each much
`smaller
`mesothelioma.
`for peritoneal
`four mutation
`rates
`and only
`
`of
`
`the framework
`Within
`these results
`the TSCE model,
`of
`that
`the higher
`suggest
`of pleural
`age-specific
`incidence
`mesothelioma
`can be mainly
`index
`of
`attributed
`to a higher
`initiation
`r, due
`either
`to a higher
`background
`diversity
`y, or a lower
`The
`rate
`of
`cell
`±, or both.
`rate,
`division,
`initiation
`rate, v, depends
`both on the number
`of
`target
`cells
`and on the rate of
`the initiating
`mutation.
`For peritoneal
`the age-specific
`incidence
`mesothelioma,
`in Fig.
`2 can be interpreted
`curves
`to be the best estimates
`from the SEER
`data
`of
`the
`incidence
`in
`a population
`the age-
`unexposed
`to asbestos.
`For pleural mesothelioma,
`specific
`incidence
`the best estimate
`curves
`represent
`of
`the
`age distribution
`of pleural
`mesothelioma
`in a
`incidence
`population
`unexposed
`to asbestos,
`but
`the actual magnitude
`of
`the age-specific
`incidence
`is not
`rates
`identifiable.
`
`Temporal
`
`trends
`
`of
`
`the period
`Over
`have been no secular
`there
`this
`study,
`in peritoneal
`among women
`and only
`mesothelioma
`trends
`weak
`observation
`that
`This
`trends
`among men.
`suggests
`for only a minor
`fraction
`asbestos
`exposure was responsible
`1973-
`of peritoneal
`mesotheliomas
`in SEER over
`the period
`2005. Spirtas
`et al.
`reported
`that about 58% of peritoneal
`[6]
`mesotheliomas
`among men in their
`population
`were
`study
`attributable
`to asbestos
`exposure.
`For
`they were
`females,
`unable
`to estimate
`separate
`for pleural
`attributable
`fractions
`the attrib-
`and peritoneal
`but
`reported
`that
`mesotheliomas,
`utable
`fraction
`for both sites combined
`was 23%. Our
`results
`here suggest
`in the SEER data over
`that, at least
`the period
`of
`the attributable
`fraction
`for male
`peritoneal
`observation,
`mesotheliomas
`was lower
`than that
`reported
`by Spirtas.
`Similar
`results
`have
`been
`reported
`in other
`registries.
`Hemminki
`and Li
`of
`[28] examined
`trends
`in the incidence
`1961-
`peritoneal
`mesothelioma
`in Sweden
`over
`the period
`29% had "typ-
`1998. Among
`men they
`reported
`that only
`jobs..."
`age-
`ical
`asbestos-related
`the
`Interestingly,
`adjusted
`incidence
`identical
`rates were virtually
`in men and
`to be occupa-
`women.
`Since men
`are much more
`likely
`that
`exposed
`to asbestos,
`this
`suggests
`finding
`tionally
`large fraction
`of peritoneal
`in Sweden
`over
`mesotheliomas
`this period were unrelated
`to asbestos
`exposure. Moreover,
`the generally
`trends
`in incidence
`over
`this period
`increasing
`are probably
`to factors
`other
`than asbestos.
`attributable
`Burdorf
`et al.
`examined
`the incidence
`of peritoneal
`[29]
`mesothelioma
`among men and women
`in Sweden
`and the
`Netherlands
`over
`the
`period
`1989-2003
`and
`reported
`absence
`of any trends.
`absence of a
`"[t]he
`concluded,
`They
`time trend in the incidence
`rate of peritoneal
`mesothelioma
`in Sweden
`and the Netherlands
`in the past
`15 years may
`point
`to a more
`limited
`role
`of occupational
`exposure
`to
`in the etiology
`asbestos
`of peritoneal
`than
`mesothelioma
`women."
`for pleural mesothelioma,
`among
`
`a
`
`especially
`
`Springer
`
`

`

`FILED: ALBANY COUNTY CLERK 01/17/2024 12:32 PM
`NYSCEF DOC. NO. 473
`942
`
`INDEX NO. 908796-22
`
`RECEIVED NYSCEF: 01/17/2024
`Cancer Causes Control
`(2009) 20:935-944
`
`trends
`
`of
`
`a
`
`only
`over
`
`a
`
`282
`Selikoff
`and
`Seidman
`peritoneal
`reported
`[30]
`of 453 mesotheliomas
`out of a total
`in the
`mesotheliomas
`cohort
`of US and Canadian
`expect
`to
`insulators.
`One would
`see such a large
`number
`of cases reflected
`in the temporal
`in a population-
`trends
`in peritoneal
`mesothelioma
`rates
`tem-
`based
`registry. Why
`then
`do we observe
`only weak
`poral
`trends
`in peritoneal
`mesothelioma
`incidence
`among
`men in the SEER database
`over
`the period
`The
`1973-2005?
`282
`peritoneal
`mesotheliomas
`reported
`by Seidman
`and
`Selikoff
`occurred
`over
`the
`20-year
`period
`1967-1986.
`Some fraction
`of
`these occurred
`in Canada
`and thus would
`not
`be reflected
`in
`the US
`statistics.
`some
`Moreover,
`fraction
`of
`the US cases occurred
`prior
`to 1973 and would
`thus not be reflected
`in the SEER database
`in 1973.
`starting
`the US pop-
`about
`10% of
`SEER represents
`Finally,
`only
`ulation.
`one
`would
`expect
`the
`peritoneal
`Thus,
`insulators'
`in the
`mesotheliomas
`cohort
`to have
`small
`impact
`on temporal
`in the SEER registry
`the period
`1973-2005.
`of peritoneal
`the epidemiology
`In
`a recent
`review
`there
`was
`that
`reported
`Boffetta
`mesothelioma,
`[31]
`correlation
`deaths
`from
`of
`between
`the
`fraction
`strong
`occupa-
`in cohorts
`pleural
`and peritoneal
`mesothelioma
`to high
`exposed
`levels
`of asbestos.
`This
`tionally
`finding
`suggests
`that
`occupational
`exposure
`to
`asbestos
`can
`mesotheli-
`the risk
`of both
`pleural
`and peritoneal
`increase
`oma.
`In contrast,
`between
`Boffetta
`reported
`low correlation
`the
`of pleural
`and peritoneal
`incidence
`mesothelioma
`in
`population-based
`This
`suggests
`in
`registries.
`that,
`finding
`fraction
`the general
`a smaller
`of peritoneal
`than
`population,
`pleural mesothelioma
`is attributable
`to asbestos
`exposure.
`As
`reported
`by Price
`and Ware
`trends
`in pleural
`[8],
`mesotheliomas
`among men clearly
`reflect
`temporal
`trends
`cohort-
`in asbestos
`use in the US. There
`has been a strong
`wise increase
`in the rates of pleural mesotheliomas
`among
`men
`the early
`a peak with
`the birth
`cohorts
`of
`reaching
`of 1965
`1920s
`thereafter.
`and declining
`The
`birth
`cohort
`appears
`to have
`the
`same
`risk
`as that
`of
`approximately
`so that
`the
`cohort
`of
`1965
`the
`epidemic
`of
`1890,
`asbestos-induced
`pleural
`mesothelioma
`appears
`to have
`abated. Among
`cohort
`the birth
`effects
`appear
`to
`women,
`have
`albeit much
`slower
`than
`the men,
`increased,
`among
`until
`about
`the cohort
`of
`birth
`1925.
`The
`effects
`cohort
`appear
`to be more
`or
`less
`flat
`from
`1925
`to
`1965.
`As
`reported
`earlier
`in 'Results',
`period
`effects
`are declining
`to the gen-
`the women,
`which
`contributes
`among
`slightly
`flat age-adjusted
`incidence
`rates despite
`the increase
`erally
`effects. We note here that
`in birth cohort
`information
`on the
`earliest
`and latest
`among men
`are based
`on few
`cohorts
`observations
`so that conclusions
`about
`them must be made
`with
`caution.
`The differences
`effects
`between
`
`by
`
`cohort
`of birth
`in the patterns
`suggest
`and
`males
`females
`
`and period
`that
`factors
`
`Springer
`
`other
`than
`the observed
`Figures
`
`exposure
`asbestos
`temporal
`trends.
`the
`show
`and
`7
`6
`and peritoneal
`pleural
`developing
`cohort
`adjusted
`and unadjusted
`respectively.
`is of
`interest
`It
`probabilities
`in males and females
`and latest birth
`cohorts,
`sugg