LORIS: A Cancer Immunotherapy Response Scoring System

LORIS is a powerful tool for predicting patient objective ICB (Immune Checkpoint Blockade) response odds and survival following ICB treatment based on common clinical, pathologic, and genomic features. It offers a user-friendly graphic user interface (GUI) for ease of use.

Figure 1. Development of the LOgistic Regression-based Immunotherapy-response Score (LORIS).

License

LORIS source code is provided under the GPLv3 license. Additionally, the trained models used by LORIS, located in this package at LORIS/model_Params, are provided under the CC BY NC 4.0 license for academic and non-commercial use.

Installation

LORIS can be installed from the github repository:

git clone https://github.com/rootchang/LORIS_GUI.git
cd LORIS_GUI
python setup.py install

Usage

Run LORIS from the command line with the following command:

LORIS

This will launch the GUI for further interaction (see Figure 2).

Figure 2. GUI of LORIS.

Examples

Example 1: Inquiry for a Single Patient

1. Input the required clinical features (e.g., TMB, albumin, NLR, and age) of the patient in the GUI (highlighted as red box 1 in Figure 2).

2. Select the desired model (e.g., LLR6_pan-cancer or LLR6_NSCLC).

3. Click the Calculate button.

4. Obtain the predicted ICB outcome of the patient, including:

   (i) Patient score (0-1) with a 95% confidence interval.

   (ii) Objective response odds (%) with a 95% confidence interval.

   (iii) Median progression-free survival (months) with a 95% confidence interval.

   (iv) Median overall survival (months) with a 95% confidence interval.

Note: Before clicking the Calculate button, please make sure to clear the contents of the Input from file entry box.

Example 2: Inquiry for Multiple Patients

1. Store the necessary clinical features of patients in a .txt file (highlighted as red box 2 in Figure 2; refer to the template file at ./LORIS/query_patients.txt for the format).

2. Load the file from the GUI.

3. Select the model (e.g., LLR6_pan-cancer or LLR6_NSCLC).

4. Click the Calculate button.

5. Retrieve the predicted ICB outcomes from the output file, which is named inputFilePrefix+"_out.txt".

Citation

If you use LORIS in your research or work, please consider citing our publication:

Tian-Gen Chang, Yingying Cao, Hannah J. Sfreddo, Saugato Rahman Dhruba, Se-Hoon Lee, Cristina Valero, Seong-Keun Yoo, Diego Chowell, Luc G. T. Morris, Eytan Ruppin. "Robust prediction of patient outcomes with immune checkpoint blockade therapy for cancer using common clinical, pathologic, and genomic features." bioRxiv 2023.07.04.547697; doi: https://doi.org/10.1101/2023.07.04.547697

Contact

For any questions or inquiries, please feel free to contact Tiangen Chang at changtiangen@gmail.com.