Skip to contents

Simulate a RECIST dataset

Source: R/data.R
example_rc.Rd

Internal function that simulates a synthetic RECIST dataset following the conventions of clinical oncology trials. In the simulation, Target Lesions response depends on the treatment arm enrolres$arm (not enrolres$arm3).

Usage

example_rc(
  enrolres,
  seed,
  rc_num_timepoints = 5,
  rc_p_new_lesions = 0.09,
  rc_p_na = 0.005,
  rc_p_nt_lesions_yn = 0.5,
  rc_p_nt_lesions_resp = list(CR = 0.73, SD = 0.25, PD = 0.01, NE = 0.01),
  rc_sd_tlsum_noise = 0.5,
  rc_coef_treatement = 3,
  ...
)

Arguments

enrolres

the enrolment result table, from example_enrol().

seed

Integer. Random seed for reproducibility (can be NULL).

rc_num_timepoints

Integer. Number of timepoints for each patient, including baseline.

rc_p_new_lesions

Integer. Probability of a new lesion

rc_p_na

Integer. Probability of a missing value in reponses of Target Lesions, Non-Target Lesions, and New Lesions independently. They add up for the global response.

rc_p_nt_lesions_yn

Integer. Probability of having Non-Target Lesions

rc_p_nt_lesions_resp

Integer list. Probability of each Non-Target Lesions response, if present

rc_sd_tlsum_noise

Integer. Standard deviation for the evolution of the Target Lesion sum of width

rc_coef_treatement

Integer. Differentiates the difference in effect between the control and treatment arm (2 arms only). For example, rc_coef_treatement = 2 mean that the growth rate of the tumor is divide per 2 and the elimination rate is multiplied per 2. Also, the probability of a new lesion is multiplied by 2.

Value

A tibble with N rows and the following columns:

  • subjid: The patient identifier

  • rcvisit: The visit number

  • rcdt: The visit date

  • rctlsum: The Target Lesion length sum at each time point. The evolution of the value is calculated based on the percentage variation in tumor size from the previous time point. This variation is simulated using a uniform distribution between -30 and 30, with added noise (the noise follows a normal distribution with a mean of 0 and a standard deviation of rc_sd_tlsum_noise).

  • rctlresp: The response associated with Target Lesions

  • rcntlresp: The response associated with Non-Target Lesions

  • rcnew: The appearance of a new lesion

  • rcresp: The global RECIST response