Skip to content

Latest commit

 

History

History

README.md

Home / Modules / Tumor

Tumor

Basal cell carcinoma (BCC) neoplastic growth model with soft contact inhibition and UWYN continuum handoff.

Biology

Basal cell carcinoma is the most common human malignancy, arising from the basal layer of the epidermis. BCC cells have an altered cell cycle (approximately 56 hours vs 17 hours for normal keratinocytes) and exhibit soft contact inhibition mediated by the YAP/TAZ mechanotransduction pathway. Unlike hard contact inhibition where proliferation stops above a neighbor threshold, soft CI produces a gradual quadratic suppression that allows continued growth at reduced rates in crowded environments.

BCC growth is slow compared to many cancers, with volume doubling times of approximately 148 days. The proliferation index (Ki-67 positive fraction) is typically around 27%, with a cell loss factor of approximately 0.112 per day from stochastic apoptosis.

Model

TumorCell agent: inherits from Cell with BCC-specific properties:

  • Cell cycle phases scaled by cycle_factor (3.3x normal)
  • Soft contact inhibition: division probability = max(0, 1 - (n_neighbors / max_neighbors)^ci_steepness)
  • Stochastic apoptosis at apoptosis_rate per step
  • No differentiation (BCC cells remain basal-like)

Seeding: at seed_time_h, a cluster of seed_count cells is placed at (seed_x, seed_y, seed_z) in a sphere distribution.

UWYN handoff: when a tumor cell enters G0 and remains quiescent for handoff_delay_h, it converts to the Tumor field (binary marker, stratum_value = 10.0) and the agent is removed. This allows the growing tumor periphery to use agents while the stable interior reverts to continuum.

Tumor PDE: binary marker field with no diffusion and no decay. Records the spatial footprint of quiescent tumor cells after handoff.

Scale-aware validation: the simulation operates at a representative tissue scale (30x30 um domain). Growth metrics are compared against clinical data using surface-to-volume ratio corrections.

Scenario composability: the tumor module can run alone (--study=tumor) or combined with wound healing (--study=tumor-wound), where a tumor grows at t=0 and a wound is introduced at day 8. Both cellular populations coexist, sharing the same field infrastructure.

Parameters

From modules/tumor/config.toml:

Parameter Default Units Description Source
enabled false bool Master switch Convention
seed_time_h 0 hours Time to seed tumor Convention
seed_x 15.0 um Cluster center X Convention
seed_y 15.0 um Cluster center Y Convention
seed_z 2.0 um Cluster center Z (basal layer) Convention
seed_count 5 count Initial cluster size Calibrated
diameter 4.0 um Cell diameter Convention
cycle_factor 3.3 multiplier BCC cycle ~56h Khoo et al. 2019 (DOI)
max_neighbors 12 count Contact inhibition threshold Convention
ci_steepness 2.0 exponent Soft CI (0=hard, 2=quadratic YAP/TAZ) Calibrated
growth_rate 5 - Volume growth rate Convention
max_cells 5000 count Carrying capacity (0 = unlimited) Convention
handoff_delay_h 20 hours G0 hours before continuum conversion Calibrated
stratum_value 10.0 - Stratum field value for tumor tissue Convention
apoptosis_rate 0.00047 per step Cell loss 0.112/day Khoo et al. 2019

Coupling

Reads

Field Source module How used
O2 tissue Proliferation gating (optional)

Writes

Field Consumer modules What is written
Tumor (visualization) Binary marker from UWYN handoff
Stratum tissue Tumor tissue value (10.0) at handoff

Validation

Dataset Observable Sources Notes
tumor_growth_rate Volume growth over time Fijalkowska 2023, Sykes 2020, Kricker 2014 Slow BCC growth
tumor_doubling_time Volume doubling time ~148 days Khoo 2019, Tejera 2023 Clinical measurements
tumor_proliferation_index Ki-67 ~27% Toth 2012, Alferraly 2019, al-Sader 1996 Immunohistochemistry

Literature data

Reference curves for validation (full citations in SOURCES.yaml):

Dataset File Normalization
Tumor doubling time tumor_doubling_time.csv Absolute days
Tumor growth rate tumor_growth_rate.csv Absolute mm
Tumor proliferation index tumor_proliferation_index.csv Absolute %
Raw digitized data (8 papers)
File Source
fijalkowska2023_bcc_growth_meta.csv Fijalkowska et al. 2023
sykes2020_superficial_bcc.csv Sykes et al. 2020
kricker2014_community_bcc.csv Kricker et al. 2014
al_qahtani2020_bcc_doubling.csv Al-Qahtani et al. 2020
tejera2023_skin_cancer_tdt.csv Tejera et al. 2023
toth2012_bcc_ki67.csv Toth et al. 2012
alferraly2019_scc_ki67.csv Alferraly et al. 2019
alsader1996_bcc_scc_indices.csv Al-Sader et al. 1996

Metrics

Column Units Description
n_tumor_cells count Active tumor cell agents
n_tumor_cycling count Tumor agents not in G0 (Ki-67 proxy)
tumor_field_cells count Tumor field voxels > 0.5 (handoff footprint)

Source files

File Purpose
tumor_cell.h TumorCell agent with BCC-specific cell cycle and soft CI
tumor_behavior.h Division, apoptosis, and UWYN handoff logic
tumor_initiation.h Sphere seeding at configured time and location
tumor_pde.h Tumor binary marker field
config.toml Module configuration