Advanced HCC: CELESTIAL Trial

Pierre Gholam, MD: Certainly if somebody is on a tyrosine kinase inhibitor [TKI] and experiences a very significant burden of adverse events, one could look a little closer at what those adverse events are and perhaps tailor a different option to the patient’s needs.

For example, if someone has a lot of hypertension or a lot of hand-foot skin reaction, and you do not feel you’re able to handle that very appropriately with dose reduction, mitigation of adverse events, and discontinuation, you might find it reasonable to switch to a different agent for that purpose.

One of these agents of course is cabozantinib. Cabozantinib is a multitarget TKI. It affects the VEGF pathways, VEGF1, VEGF2, and VEGF3. It also certainly affects other pathways, including MET and AXL, which are potentially important in preventing spread and metastases of disease and potentially are implicated in resistance to the effects of drugs such as sorafenib. Cabozantinib was studied in a large study called CELESTIAL in the second-line setting. Compared with placebo, these were patients who were able to receive sorafenib in the first line, so every patient who was in that study received sorafenib first.

But the study had fairly generous inclusion and exclusion criteria in that patients could have received a second-line agent that could include even an I/O [immuno-oncology] treatment. There was even a small number of patients who had a third-line of treatment before being enrolled in CELESTIAL. CELESTIAL basically has its data reported as of 2018 in the New England Journal of Medicine, and CELESTIAL data show that at the planned interim analysis, a median overall survival of 10.2 months with cabozantinib, and 8 months with placebo for a hazard ratio of 0.76. Remind you, this is in the second-line setting. A progression-free survival with CABO [cabozantinib] was 5.2 months and 1.9 months with placebo. This was for all comers, including patients who had multiple lines of therapy, which was a significant proportion of these patients.

The advantages and disadvantages of prescribing this drug versus other drugs in part has to do with mechanistic components, in terms of the pathways they target and how these may—at least in theory—circumvent resistance. We do not have specific clinical evidence that supports this without any shadow of a doubt, but at least mechanistically one could see how in the second line a drug that targets MET and AXL might be able to help with that.

The other aspects that are of course important are the sequencing of a TKI after initiation of therapy with another TKI and the ability to tolerate that initial treatment, which obviously is built into the inclusion criteria of CELESTIAL.

The primary efficacy data in the second line in terms of overall survival and progression-free survival have been described to you just now. In addition, it’s important to understand that in patients who had only 1 line of therapy with sorafenib—pretty much everybody had to have at least sorafenib in the first line—in a subset analysis, these increase more differentially compared with just the overall cohort of patients.

Cabozantinib, like every other drug, is not by any means free of adverse events. The adverse events that are hallmarks of TKIs are certainly present here. These include hand-foot skin reaction, hypertension, and fatigue. And these certainly need to be aggressively managed by an experienced physician who is very familiar with the adverse events of TKIs, in order to maintain patients on therapy and hopefully achieve the same type of survival benefit in the second line that a patient enrolled in the CELESTIAL trial was able to benefit from.

Hand-foot skin reaction, for anyone who is prescribed TKIs, should be a staple adverse event that we manage in early stages with low degrees of severity using topicals, such as 5% urea cream; avoiding repetitive motion in areas of high impact; and applying treatments to any calluses, corns, or other problems in the feet and the hands, which would be usually delivered by a podiatrist.

But when the severity of these toxicities is more pronounced, such as when you have partial or full thickness alteration of hands, fingers, or toes, typically dose interruption or dose discontinuation for an interval period of time until the subset of toxicities subside is the way to go. In some cases, if that does not resolve, discontinuation of the drug is really the only way to do this.

Hypertension is typically managed with an antihypertensive, sometimes with dose reduction or dose discontinuation if this does not work. I might go antihypertensive in the setting of cabozantinib treatment as a calcium channel blocker, either initiation or an increase in dose. If the patient is already hypertensive, then I would say that many patients are already hypertensive when they have these toxicities because they have it set up for them. Fatigue is overwhelming at times, and it affects all TKIs. So if you are treating a patient with a TKI, you have to be able to manage this effectively. At times with dose reductions from let’s say 60 to 40 in the case of cabozantinib. At other times you need to dose interrupt in order to allow the patient to recover.

In the setting of liver disease in general, and HCC as well, increasing physical activity seems to be an effective treatment for fatigue in some patients. Certainly identifying any mental health burdens that could be overlying this, such as depression, is important to diagnose and treat. There have been some small trials looking at other interventions for fatigue, which I very selectively offer patients, including acupuncture and other alternative and complementary therapies.

There are certainly some patients who have a high AFP [alpha-fetoprotein], where we see some tyrosine kinase inhibitor treatments do less well. That does not appear to be so much the case with cabozantinib, which is a good thing. I’ll remind this audience that there is a treatment in the second line, specifically approved for patients who have an AFP greater than 400 ng/mL, and that’s ramucirumab. That restriction in approval does not apply, though, to cabozantinib, which is approved at all AFP levels. But it certainly is something to think about and follow over time in those patients.

Transcript edited for clarity.

Case: A 61-Year-Old Man with Stage 4 Hepatocellular Carcinoma

Initial presentation

• A 61-year-old man presented with nausea, vomiting, decreased appetite and occasional generalized itching
• PMH: diabetes, medially controlled; hepatitis C and B coinfection diagnosed and treated 6 years ago; diagnosed with HCC December 2018
• PE: scleral icterus; jaundice; spleen palpable 4-cm below the costal margin
  
  

Clinical workup

• Labs: AFP 436 ng/mL, bilirubin 1.6 mg/dL, AST 105 U/L, ALT 110 U/L, ALP 390 U/L, INR 1.9, albumin 3.8 g/dL, BUN 13 mg/dL, creatinine 1 mg/dL, plt 95,000
• HBV+, HCV+
• Abdominal ultrasound revealed 3 small hepatic lesions
• Chest/abdominal/pelvic CT scan confirmed 2 focal nodules in the right and 1 in the left hepatic lobe measuring 3.6 cm, 4.9 cm and 5.2 cm, a suspicious lesion in the right lower lung lobe; wide-spread lymphadenopathy was noted
• Biopsy findings showed grade 3 hepatocellular carcinoma with marked fibrosis
• Surgical consult: unresectable due to tumor size and location
• Child-Pugh A; BCLC stage C
• ECOG 1
  
  

Treatment and Follow-Up

• 2018: treated with lenvatinib 400 mg PO q12hr; he experienced diarrhea for 2 weeks which resolved; achieved PR
• 2020: Imaging showed a new lung lesion
  • Treatment was with cabozantinib 60 mg PO qDay was initiated