Genomic adjusted radiation dose

  1. Home
  2. Wiki
This article is about the genomics-based dose-effect model. For the biological concept, see Radiogenomics.

Genomic adjusted radiation dose (GARD) is a framework in radiation oncology that estimates the biological effect of a given physical radiation dose by combining a tumor's gene-expression–derived radiosensitivity with a radiobiological dose–effect model (e.g. the linear–quadratic model).

Radiation sensitivity index (RSI)

[edit]
Main article: Radiation sensitivity index

The Radiation Sensitivity Index (RSI) is a gene-expression–based model developed to estimate intrinsic tumor cell radiosensitivity. RSI was originally derived by correlating cancer cell line survival after radiation with baseline gene-expression patterns, and has been validated in multiple retrospective clinical cohorts. RSI serves as the genomic basis for the Genomic Adjusted Radiation Dose (GARD) framework.

Background

[edit]

Conventional radiotherapy is typically prescribed using fixed schedules (e.g. 2 Gy per fraction) that do not account for inter-tumor variability in radiosensitivity.[1] Advances in genomic profiling and radiogenomic research have led to efforts to model how gene-expression patterns influence tumor response to radiation. Among these, the Genomic Adjusted Radiation Dose (GARD) framework was proposed to quantify the expected biological effectiveness of a given physical dose for an individual tumor, enabling genomically informed dose personalization.[2][3]

Other methods to predict radiosensitivity have also been explored. These include integrative radiogenomic models that correlate tumor gene-expression with in vitro radiosensitivity,[4] imaging-based proxies such as FDG-PET–derived voxel dose–response mapping using serial PET/CT feedback imaging,[5][6] and mathematical frameworks such as the Proliferation–Saturation Index (PSI) and Dynamics-Adapted Radiotherapy Dose (DARD).[7][8] Many of these approaches—including GARD—have primarily been evaluated in retrospective or observational settings, and prospective validation studies are ongoing.[9]

Origins and methodology

[edit]

Radiosensitivity Index (RSI)

[edit]

The foundation of GARD is the Radiosensitivity Index (RSI), derived from a 10-gene expression model trained to predict surviving fraction at 2 Gy (SF₂) in cell lines.[10] Subsequent work refined the model via systems-biology/network modeling in two companion International Journal of Radiation Oncology • Biology • Physics papers in 2009.[11][12]

Genomic Adjusted Radiation Dose (GARD)

[edit]

GARD integrates RSI with the linear–quadratic formalism to estimate the biological effect of a given physical dose for an individual tumor was first introduced in 2017.[2] A pooled pan-cancer analysis later examined GARD in multiple tumor types.[13]

Evidence and validation

[edit]

RSI-only validation (pre-2017)

[edit]

Following development of the RSI, several studies assessed its prognostic and predictive utility in human tumors. An early clinical validation in Breast cancer demonstrated that RSI was associated with clinical outcomes among patients receiving radiotherapy.[14] Subsequent disease-specific analyses showed that RSI predicted overall survival in glioblastoma[15] and in pancreatic cancer patients receiving adjuvant radiotherapy.[16]

GARD-based validation (2017—present)

[edit]

A pooled multi-cohort analysis across several cancers reported that GARD was associated with benefit from radiotherapy when analyzed alongside conventional dose metrics.[13] Disease-specific applications include triple-negative breast cancer,[17] lung metastases treated with stereotactic body radiotherapy,[18] and HPV-positive oropharyngeal cancer (OPSCC).[19] The body of literature to date is largely retrospective or observational; prospective evaluation is ongoing.

Limitations and future directions

[edit]

While GARD has demonstrated reproducible associations with radiotherapy outcomes across multiple cancers, several areas deserve continued refinement as the field moves toward personalized radiation dosing. Tumor heterogeneity, sampling bias, and variation in oxygenation and hypoxia distribution remain important considerations, as a single biopsy may not fully represent subclonal diversity or microenvironmental gradients that influence radiosensitivity.[20][21][22] Classical tumor control probability (TCP) models also emphasize that dose–response relationships depend on tumor size, clonogen number, and spatial cell distribution, parameters not explicitly incorporated in current GARD formulations.[23] Recent developments in imaging-based biomarkers, including radiomics and voxel-level dose–response mapping, offer complementary ways to characterize tumor biology and spatial heterogeneity that could further inform GARD-based planning.[24] Ongoing work is focused on integrating genomic, imaging, and spatial data and on prospective and real-world evaluation to enhance the precision and generalizability of biologically informed radiotherapy dosing.

Ongoing clinical trials

[edit]
  • NCT05528133 – Genomically Guided Radiation Therapy in Triple-Negative Breast Cancer (feasibility).[25]
  • NCT05873439 – Genomically Guided Radiation Dose Personalization in Locally Advanced NSCLC (feasibility).[26]

See also

[edit]

References

[edit]
  1. ^ Harary, PM (2024). "Genomic predictors of radiation response: recent progress". Cell Death & Disease. 15 (1): 376. doi:10.1038/s41420-024-02270-2. PMC 11213856. PMID 38942810.
  2. ^ a b Scott, JG (2017). "A genome-based model for adjusting radiotherapy dose (GARD): a retrospective, cohort-based study". Lancet Oncology. 18 (2): 202–211. doi:10.1016/S1470-2045(16)30648-9. PMC 7771305. PMID 27993569.
  3. ^ Fillon, M (2022). "Genomic-derived radiation dosage improves prediction of patient benefit". CA: A Cancer Journal for Clinicians. 72 (4): 305–307. doi:10.3322/caac.21711. PMID 34874554.
  4. ^ Abazeed, ME (2013). "Integrative radiogenomic profiling of squamous cell lung carcinoma". Cancer Research. 73 (20): 6289–6298. doi:10.1158/0008-5472.CAN-13-1616. PMC 3856255. PMID 23980093.
  5. ^ Yan, D (2019). "Tumor Voxel Dose-Response Matrix and Dose Prescription Function Derived Using 18F-FDG PET/CT Images for Adaptive Dose Painting by Number". International Journal of Radiation Oncology • Biology • Physics. 104 (1): 207–218. doi:10.1016/j.ijrobp.2019.01.077. PMID 30684661.
  6. ^ Chen, S (2022). "Dynamic Characteristics and Predictive Capability of Tumor Voxel Dose-Response Assessed Using 18F-FDG PET/CT Imaging Feedback". Frontiers in Oncology. 12 876861. doi:10.3389/fonc.2022.876861. PMC 9299377. PMID 35875108.
  7. ^ Sunassee, E (2019). "Proliferation Saturation Index in an adaptive Bayesian framework for personalised radiotherapy". Radiation Oncology. 14 (10): 1421–1426. doi:10.1080/09553002.2019.1589013. PMC 7081883. PMID 30831050.
  8. ^ Zahid, MU (2021). "Dynamics-Adapted Radiotherapy Dose (DARD) for head and neck cancer radiotherapy dose personalization". Frontiers in Oncology. 11: 784039. doi:10.3389/fnbeh.2021.777778. PMC 8689317. PMID 34938167.
  9. ^ Yin, J (2025). "Narrative Review of the Use of Genomic-Adjusted Radiation Dose (GARD) in Radiotherapy". Cancers (Basel). 17 (3): 452–459. doi:10.3390/cancers17030845 (inactive 4 October 2025). PMC 12384496. PMID 39816421.{{cite journal}}: CS1 maint: DOI inactive as of October 2025 (link)
  10. ^ Torres-Roca, JF (2005). "Prediction of radiation sensitivity using a gene expression classifier". Cancer Research. 65 (16): 7169–7176. doi:10.1158/0008-5472.CAN-05-0656. PMID 16103067.
  11. ^ Eschrich, SA (2009). "A gene expression model of intrinsic tumor radiosensitivity: prediction of response and prognosis after chemoradiation". International Journal of Radiation Oncology • Biology • Physics. 75 (2): 489–496. doi:10.1016/j.ijrobp.2009.04.050. PMC 3038688. PMID 19735873.
  12. ^ Eschrich, S (2009). "Systems biology modeling of the radiation sensitivity network: a biomarker discovery platform". International Journal of Radiation Oncology • Biology • Physics. 75 (2): 497–505. doi:10.1016/j.ijrobp.2009.04.038. PMC 2762403. PMID 19735874.
  13. ^ a b Scott, JG (2021). "Pan-cancer prediction of radiotherapy benefit using genomic-adjusted radiation dose (GARD): a cohort-based pooled analysis". Lancet Oncology. 22 (9): 1221–1229. doi:10.1016/S1470-2045(21)00347-8. PMID 34363761.
  14. ^ Eschrich, SA (2012). "Validation of a Radiosensitivity Molecular Signature in Breast Cancer". Clinical Cancer Research. 18 (18): 5134–5143. doi:10.1158/1078-0432.CCR-12-0891. PMC 3715399. PMID 22872574.
  15. ^ Ahmed, KA (2015). "The radiosensitivity index predicts for overall survival in glioblastoma". Oncotarget. 6 (33): 34414–34422. doi:10.18632/oncotarget.5437. PMC 4741462. PMID 26451615.
  16. ^ Strom, T (2015). "Radiosensitivity index predicts for survival with adjuvant radiation in resectable pancreatic cancer". Radiotherapy and Oncology. 117 (1): 159–164. doi:10.1016/j.radonc.2015.07.018. PMC 7771365. PMID 26235848.
  17. ^ Ahmed, KA (2019). "Utilizing the genomically adjusted radiation dose (GARD) to personalize adjuvant radiotherapy in triple-negative breast cancer management". eBioMedicine. 47: 163–169. doi:10.1016/j.ebiom.2019.08.025. PMC 6796554. PMID 31462392.
  18. ^ Ahmed, KA (2018). "Radiosensitivity of lung metastases by primary histology and implications for stereotactic body radiation therapy using the genomically adjusted radiation dose". Journal of Thoracic Oncology. 13 (8): 1121–1127. doi:10.1016/j.jtho.2018.04.027. PMC 7810135. PMID 29733909.
  19. ^ Ho, E (2025). "Personalized treatment in HPV + oropharynx cancer using genomic adjusted radiation dose". Journal of Clinical Investigation. 135 (19) e194073. doi:10.1172/JCI194073. PMC 12483556. PMID 40996827.
  20. ^ Kashyap, A (2022). "Quantification of tumor heterogeneity: from data acquisition to modeling". Trends in Biotechnology. 40 (3): 309–324. doi:10.1016/j.tibtech.2021.05.012 (inactive 4 October 2025).{{cite journal}}: CS1 maint: DOI inactive as of October 2025 (link)
  21. ^ Buffa, F (2010). "Large meta-analysis of multiple cancers reveals a common, compact and highly prognostic hypoxia metagene". British Journal of Cancer. 102 (2): 428–435. doi:10.1038/sj.bjc.6605450. PMC 2816644. PMID 20087356.
  22. ^ Scott, JG (2016). "Spatial metrics of tumour vascular organisation predict radiation efficacy in a computational model". PLOS Computational Biology. 12 (1) e1004712. Bibcode:2016PLSCB..12E4712S. doi:10.1371/journal.pcbi.1004712. PMC 4723304. PMID 26800503.
  23. ^ Spoormans, K (2022). "A review on tumor control probability (TCP) and its applications in radiotherapy". Frontiers in Oncology. 12 847295. doi:10.3389/fonc.2022.847295 (inactive 4 October 2025).{{cite journal}}: CS1 maint: DOI inactive as of October 2025 (link)
  24. ^ Lou, B (2019). "An image-based deep learning framework for individualizing radiotherapy dose". The Lancet Digital Health. 1 (3): e25 – e34. doi:10.1016/S2589-7500(19)30058-5. PMC 6708276. PMID 31448366.
  25. ^ "Study of Genomically Guided Radiation Therapy in Triple Negative Breast Cancer (NCT05528133)". ClinicalTrials.gov. 2025-09-09. Retrieved 2025-10-04.
  26. ^ "Study of Genomically Guided Radiation Dose Personalization in NSCLC (NCT05873439)". ClinicalTrials.gov. Retrieved 2025-10-04.