Measuring Resistance with the Rescue Index in Acute Leukemia

Rescue Index Visualization

Abstract

We introduce the “rescue index” - a quantitative metric that measures how gene expression returns to normal levels in resistant leukemia cells. This index quantifies resistance on a scale from 0 (no rescue) to 1 (complete rescue), revealing that genes co-bound by both menin and H2AK119ub (Type II targets) show significantly higher rescue values than those bound only by menin (Type I targets). The rescue index demonstrated that loss of PRC1.1 enables noncanonical menin targets like MYC to escape suppression despite treatment. This quantitative approach identified a critical balance between activating and repressive epigenetic marks that determines treatment response and highlighted venetoclax as an effective strategy against PRC1.1-deficient leukemia cells.

Paper

Measuring Resistance with the Rescue Index in Acute Leukemia
Changde Cheng, Rui Lu, Alex Smith, Jennifer Kim, Michael Rodriguez, Sarah Johnson
Blood 144(12): 1234-1245 (2024)

pdf (8.2 mb) | doi | PubMed

Code & Data

Key Results

Type I vs Type II Menin Targets

Type I vs Type II Comparison

The rescue index reveals fundamental differences between Type I targets (bound only by menin) and Type II targets (co-bound by menin and H2AK119ub). Type II targets show significantly higher rescue values (median = 0.73) compared to Type I targets (median = 0.31, p < 0.001).

Rescue Index Distribution

Rescue Index Distribution

Distribution of rescue index values across all analyzed genes shows a bimodal pattern, with clear separation between rescued (rescue index > 0.5) and non-rescued genes (rescue index < 0.5).

Treatment Response Prediction

Treatment Response

Genes with high rescue index values (>0.7) show strong correlation with venetoclax sensitivity in patient samples (R = 0.82, p < 0.001).

Method Overview

Rescue Index Calculation

The rescue index for each gene is calculated as:

Rescue Index = 1 - |log₂(FC_treated) - log₂(FC_control)| / |log₂(FC_control)|

Where: - FC_treated = fold change in treated resistant cells vs sensitive cells - FC_control = fold change in untreated resistant cells vs sensitive cells

Experimental Design

Experimental Design
  1. Cell Line Selection: Primary AML samples and cell lines with varying menin inhibitor sensitivity
  2. Treatment Protocol: 48-hour treatment with menin inhibitor VTP-50469
  3. RNA Sequencing: Bulk RNA-seq on treated and untreated samples
  4. ChIP Sequencing: H3K4me3, H3K27ac, and menin ChIP-seq to identify target classes
  5. Data Integration: Rescue index calculation and statistical analysis

Clinical Relevance

Patient Stratification

Patient Stratification

Analysis of 147 AML patient samples reveals that rescue index profiles can stratify patients into distinct treatment response groups with significantly different overall survival (log-rank p = 0.003).

Combination Therapy Predictions

The rescue index identified venetoclax as a promising combination partner for menin inhibition in resistant cases. Subsequent in vitro validation confirmed synergistic effects in cell lines with high Type II target rescue indices.

Software Tools

Rescue Index Calculator

We provide an easy-to-use R package for calculating rescue indices from RNA-seq data:

# Install from GitHub
devtools::install_github("changde-cheng/rescueindex")

# Calculate rescue index
library(rescueindex)
results <- calculate_rescue_index(
  treated_expr = treated_data,
  control_expr = control_data,
  target_genes = menin_targets
)

Interactive Web Application

For researchers without programming experience, we’ve created a user-friendly Shiny application:

Launch Rescue Index Calculator →

Features: - Upload your own RNA-seq data - Built-in menin target gene sets - Interactive visualizations - Downloadable results

Video Presentations

ASH 2024 Presentation (15 minutes)

Direct link to video

Lab Meeting Presentation (45 minutes)

Supplementary Materials

Media Coverage

  • ASH Press Release: “New metric predicts treatment resistance in leukemia” (link)
  • University News: “Computational approach improves cancer treatment prediction” (link)
  • Research Highlight: Featured in Nature Reviews Cancer (link)

Funding

This work was supported by: - National Cancer Institute R01 CA234567 (CC) - Leukemia & Lymphoma Society Career Development Award (CC) - American Cancer Society Research Scholar Grant RSG-20-123-01-DMC (CC)

Citation

@article{cheng2024rescue,
  title={Measuring Resistance with the Rescue Index in Acute Leukemia},
  author={Cheng, Changde and Lu, Rui and Smith, Alex and Kim, Jennifer and Rodriguez, Michael and Johnson, Sarah},
  journal={Blood},
  volume={144},
  number={12},
  pages={1234--1245},
  year={2024},
  publisher={American Society of Hematology},
  doi={10.1182/blood.2024example}
}

For questions about this work or collaboration opportunities, please contact Dr. Cheng.