seerSurv

Overview

seerSurv provides a transparent, reproducible framework for deriving monthly cancer-recurrence transition probabilities used in state-transition health economic models.

Given five-year aggregate survival proportions for local/regional recurrence (LR) and distant recurrence (DR) populations—typically extracted from the SEER-17 registry—the package estimates:

Symbol	Description
P_LL	Monthly probability of remaining in LR
P_DL	Monthly probability of progressing from LR to DR
P_death given LR	Monthly probability of dying in LR (1 - P_LL - P_DL)

Bundled data cover 11 cancer types (Breast, Bladder, Colon & Rectum, Esophageal, Kidney, Liver, Lung, Melanoma, Ovarian, Pancreatic, Stomach).

Installation

# Install the released CRAN version (once available)
install.packages("seerSurv")

# Install the development version from GitHub
# install.packages("pak")
pak::pak("heorlytics/seerSurv")

Quick start

library(seerSurv)

data(lifetable_seer)

result <- run_tumour_analysis(
  tumour       = "Melanoma",
  surv_vec_LR  = c(1, 0.9959, 0.9886, 0.9831, 0.9783, 0.9754),
  surv_vec_DR  = c(1, 0.5491, 0.4396, 0.3945, 0.3630, 0.3410),
  N_L          = 64775L,
  N_D          = 3121L,
  prop_male_LR = 0.580,
  mean_age_LR  = 61,
  prop_male_DR = 0.692,
  mean_age_DR  = 62,
  lifetable    = lifetable_seer,
  scenario     = "lifetime",   # "lifetime", "20y", or "5y"
  criterion    = "AIC"         # "AIC" or "BIC"
)

result
#> # A tibble: 1 x 7
#>     P_LL P_DL_approach2 P_Death_L_approach2 RMST_L_years RMST_D_years M_years M_months
#>    <dbl>          <dbl>               <dbl>        <dbl>        <dbl>   <dbl>    <dbl>
#> 1  0.999         0.0003               4e-04         17.2         3.15    14.1     169.

Multi-tumour batch analysis

library(dplyr)
library(tidyr)

data(tumour_data_seer)

results <- tumour_data_seer |>
  group_by(Tumor) |>
  summarise(
    out = list(
      run_tumour_analysis(
        tumour       = first(Tumor),
        surv_vec_LR  = c(S0[Recurrence == "LR"], S1[Recurrence == "LR"],
                         S2[Recurrence == "LR"], S3[Recurrence == "LR"],
                         S4[Recurrence == "LR"], S5[Recurrence == "LR"]),
        surv_vec_DR  = c(S0[Recurrence == "DR"], S1[Recurrence == "DR"],
                         S2[Recurrence == "DR"], S3[Recurrence == "DR"],
                         S4[Recurrence == "DR"], S5[Recurrence == "DR"]),
        N_L          = first(N_LR),
        N_D          = first(N_DR),
        prop_male_LR = prop_male[Recurrence == "LR"],
        mean_age_LR  = mean_age[Recurrence == "LR"],
        prop_male_DR = prop_male[Recurrence == "DR"],
        mean_age_DR  = mean_age[Recurrence == "DR"],
        lifetable    = lifetable_seer,
        scenario     = "lifetime",
        criterion    = "AIC"
      )
    ),
    .groups = "drop"
  ) |>
  unnest(out)

Methodology

Input: 5-year aggregate survival (LR and DR) + demographics
     |
     v
 prep_ipd()           <- convert to pseudo-IPD (probability-weighted)
     |
     v
 fit_models()         <- 6 parametric distributions (flexsurv)
     |
     v
 extract_ic() +       <- select top-k by AIC / BIC
 compute_weights()    <- assign relative-likelihood weights
     |
     v
 blend_survival()     <- weighted-average survival curve
     |
     v
 make_background_surv() <- multiply by US-CDC life-table survival
     |
     v
 run_tumour_analysis()  <- optimise P(LL) and P(DL), return tibble

P(LL) is derived from the mean LR sojourn time (RMST differential). P(DL) is estimated by minimising the relative sum-of-squares between the observed LR survival curve and its convolution approximation.

Full details are provided in the companion vignette and the associated publication.

Bundled data

Dataset	Description
`lifetable_seer`	US-CDC annual death rates by age and sex (ages 0-100)
`tumour_data_seer`	SEER-17 5-year survival proportions for 11 cancer types

Citation

If you use seerSurv in your research, please cite:

Mansoori S, Pandey S, Rani R, Singh B, Kurt M (2025). “Deriving Cancer Progression Rates After Local/Regional Recurrence Using Aggregate Survival Data: An R Package for Health Economic Evaluations.” Value in Health (submitted).

citation("seerSurv")

Contributing

Bug reports and pull requests are welcome on GitHub. Please read the contribution guidelines before opening a pull request.