Using Select Assessment Criteria May Change Prognoses in Advanced HCC

Jordi Rimola, MD, PhD

Jordi Rimola, MD, PhD

A combination of real-world data, clinical trial data, and radiologic imagery signals that using more precise assessment criteria may change prognoses for patients with advanced HCC. During the 2019 International Liver Cancer Association (ILCA) Annual Conference, Jordi Rimola, MD, PhD, presented his therapy on HCC heterogeneity with the use of conventional and functional radiologic imaging.

We observed that patients treated with nivolumab (Opdivo), a monoclonal antibody, have more heterogeneity in radiological response than patients treated with other drugs, Rimola stated. To better understand why this occurs, a retrospective analysis of data from 5 referral centers in Spain was conducted to review radiological responses in patients with HCC treated with nivolumab.

Separate assessments were run using RECIST 1.1, original RECIST, and immune RECIST (iRECIST). The analysis showed that some patients had intralesional heterogeneity, which causes some lesions to remain stable while others increase or decrease in size. Additionally, the investigators noticed a difference in the number of radiographic responses depending on which criteria was used. Specifically, more responses were seen when more lesions were targeted and original RECIST showed the highest number of responses.

In an interview withTargeted Oncologyat the ILCA conference, Rimola, radiologist, Radiology Department, Barcelona Clinic Liver Cancer (BCLC) Group, University of Barcelona, explained how assessment criteria may change patterns of progression and ultimately change outcomes in patients with advanced HCC treated with nivolumab.

TARGETED ONCOLOGY:Can you discuss the rationale for your retrospective analysis?

Rimola: The rationale for conducting this study was just to review what's happening in real life in patients with advanced HCC who underwent [treatment with] nivolumab. We had the impression that the radiological assessment of these patients is more complex than with other drugs because there is a lot intra-patient heterogeneity. This heterogeneity is higher than what we can expect with other treatments, which creates additional complexity. We [suspect] that the criteria used in other treatments for advanced HCC cannot work properly in that scenario. Also, we were interested in knowing what happens after the progression. We know those progressions can occur in different patterns, such as an increasing number of lesions, an increase in lesion size, present nominal lesions, or extrahepatic spread.

From past experiences, we saw that different progression patterns may have different meanings in terms of prognosis. We want to analyze what happened here.

TARGETED ONCOLOGY: What was the method of design for your analysis?

Rimola: We started to collect data from different referal sites in Spain, and we captured the real-life experience, clinical data, and radiologic images. I was charged with retrospectively reviewing the radiologic response of all patients that enrolled in this study.

TARGETED ONCOLOGY: What were the study findings?

Rimola: The findings were that some patients had individual heterogeneity. [This means] that one group of lesions increased in size, some remained stable, and others decreased in size. The other unexpected feature that we analyzed was the radiologic response using different criteria. [These criteria included] RECIST 1.1, which is the most commonly used in clinical trials, the original RECIST, which includes a wider number of target measurable lesions (up to ten), and finally, iRECIST, which is the RECIST adapted to immune checkpoint inhibitors. We saw that there was disagreement in the number of different radiologic responses. With [either] partial response, stable disease, and progressive disease, the number changes depending on the criteria.

After the first cycle of nivolumab, among the 22 subjects of our series, we observed an heterogeneous evolutionary pattern in tumor burden: 4 subjects presented decrease in tumor burden, although 1 patients finally died due to disease progression; tumor burden increased in 12

subjects, but in 3 of them tumor burden decreased thereafter; and finally 5 patients presented a <5% size change. This hetergenous evolution was also observed after the detection of new lesions, which in some patients was followed by a decrease in tumor burden. We needed to include more patients with our data to see what happened and that is what we did after this initial work. The work we presented at ILCA has a limited number of patients, but after this first analysis, we included more patients and more assessments so that we can analyze these data and see what happens. Maybe we will define different patterns of progression and link these patterns to the outcomes of patients. [We hope to] better identify the worst scenario and the best scenario. When a patient progresses in clinical trials, they are excluded from the study, but maybe what happens afterward is important. After initial progressions, depending on the type of progression, the patient may have a chance to continue the treatment and benefit, [otherwise] we know that it's a bad progression.

TARGETED ONCOLOGY: What is the key message that community oncologists should take away from this study.

Rimola: There is a lot of intra-patient heterogeneity. We have to look at all lesions, not just at a limited number of lesions. Assessment by RECIST 1.1 may be a limitation. We have to open our mind to include a larger number of lesions and link the pattern of progression with the outcomes.

Reference:

Rimola J, Da Fonseca LG, Sapena V, et al. Radiological response of hepatocellular carcinoma patients treated with nivolumab in real life. Multicentric analysis from referral centers in Spain. Presented at: 13th Annual Conferences of the International Liver Cancer Association; September 20-22, 2019; Chicago, IL. Abstract P-047.