Updates to the 2024 LI-RADS Treatment Response Assessment Algorithm

After six years of using the 2018 LI-RADS Treatment Response Assessment (LI-RADS TRA) algorithm, multiple questions and limitations have emerged. Consequently, evidence-based studies have been conducted for validation. Following this research, several key statements can be made:

• APHE is the only imaging feature that has demonstrated a true diagnostic yield in confirming viability.
• The LI-RADS algorithm has shown good reliability in classifying patients treated with locoregional therapies into viable and non-viable categories. However, the LR-TR Equivocal category has poor interobserver agreement, and studies have demonstrated that a significant number of patients classified as Equivocal had an incomplete response on pathological examination.
• MRI features such as T2 hyperintensity and diffusion restriction have improved sensitivity (ancillaru feautures, AF) for detecting viable lesions when combined with other imaging findings.
• Arterial mass-like hyperenhancement is a common finding in patients treated with radiation therapy. This can persist for up to 3–6 months, increasing the number of patients incorrectly classified as LR-TR Viable.

Changes in the 2024 LI-RADS TRA Algorithm

In response to these limitations, the LI-RADS Committee has introduced the following changes in the 2024 version:

• Washout and enhancement similar to pretreatment are no longer considered viability characteristics.
• AF can now be used to upgrade patients from LR-TR Equivocal or LR-TR Non-progressing to LR-TR Viable, increasing sensitivity and reducing the number of patients classified as Equivocal.
• Two separate cores for treatment evaluation have been introduced:
  • One for patients treated with locoregional therapies.
  • One for patients treated with radiation therapies.
• The LI-RADS TR radiation core incorporates a watch-and-wait strategy.

LI-RADS Non-Radiation TRA Version 2024 (Figure 6)

This version must be used for lesions treated with surgery or locoregional therapies. There are four main categories:

• LR-TR Non-evaluable: The presence of mass-like enhancement cannot be assessed due to image degradation or omission.
• LR-TR Nonviable: No mass-like enhancement (of any degree, in any phase) is identified within the treated lesion or along its margins. Normal expected findings along the margins of the treated lesion now have a clearer definition (Figures 7,8 and 9):
  • Smooth perilesional enhancement.
  • Perfusional parenchymal changes without mass-like enhancement.
• LR-TR Viable: A mass-like enhancement (of any degree, in any phase) is classified as a viable lesion. Washout and enhancement similar to pretreatment are no longer criteria for viability. Mass-like enhancement includes nodular, irregular, or smooth enhancement within the treated lesion or along its margins (Figure 10).
• LR-TR Equivocal: This category is reserved for small enhancing areas that may harbour viable tumour but are not large enough to be confidently diagnosed (Figure 11). Most of these will eventually be diagnosed as LR-TR Viable on follow-up studies (Figure 12). The current definition for LR-TR Equivocal is uncertainty about mass-like enhancement (its presence or morphology). The 2018 definition (enhancement atypical for treatment-specific expectations) was found to be too subjective and inconclusive.

In these patients, short-interval follow-up with imaging every three months is recommended. Ancillary features (T2 hyperintensity and restricted diffusion) can optionally be used to upgrade patients to LR-TR Viable.

LI-RADS Radiation TRA Version 2024 (Figure 13)

This version must be used for lesions treated with radiation therapies, such as TARE and SBRT. There are four main categories:

• LR-TR Non-evaluable: The presence of mass-like enhancement cannot be assessed due to image degradation or omission. The definition remains the same as in LI-RADS Non-Radiation.
• LR-TR Nonviable: No mass-like enhancement (of any degree, in any phase) is identified within the treated lesion or along its margins. Expected findings along the margins of the treated lesion are the same as those in non-radiation LI-RADS (Figures 14, 15, and 16):
  • Smooth perilesional enhancement
  • Perfusion-related parenchymal changes secondary to radiation:
    • After TARE, perfusional changes typically exhibit a wedge-shaped morphology, resembling ablation segmentectomy.
    • After SBRT, these changes surround the treated tumor.
  • LR-TR Viable: Mass-like enhancement (of any degree, in any phase) within the treated lesion or along its margins that is new or has increased in size over time after radiation therapy.Most patients will require at least two follow-up studies before being classified in this category. Pseudoprogression must be assessed before confirming viability.
  • LR-TR Non-progressing: Mass-like enhancement (of any degree, in any phase) within the treated lesion or along its margins that remains stable or decreases in size over time after radiation therapy. This category is reserved for treated lesions with persistent tumor enhancement due to non-viable residual tumor cells (Figure 16). These non-viable cells do not replicate and will eventually disappear. This category helps reduce false-positive viable lesions and minimizes overtreatment and its associated morbidity.

In these patients, a follow-up imaging study is recommended one to three months after treatment. Subsequent CT or MRI scans should be performed every three months.