FRESCO positive efficacy, favourable safety
HCM and partner Lilly are developing fruquintinib as a potential best-in-class drug for it to compete in the global setting. In our view, FRESCO underpins the hypothesis that fruquintinib’s safety and efficacy profile could enable it to be positioned as best in class in China. Its profile internationally will be determined by global Phase III trials; however, data from FRESCO bode well.
Full data were presented at ASCO from HCM’s first pivotal Phase III trial (FRESCO), a China-based study, which evaluated fruquintinib, a VEGFR 1/2/3 inhibitor in third-line colorectal cancer patients. Colorectal cancer is believed to be one of the five most common cancers in both Chinese men and women, with an estimated 191.0 thousand deaths in China in 2015 (incidence of 376.3 thousand in 2015). Most Chinese CRC patients are treated with chemotherapy and few targeted therapies are approved, with limited salvage treatments in third line and above. If approved, fruquintinib could capture a significant portion of the market in these patients.
The FRESCO study randomised 416 patients (519 screened) 2:1 into fruquintinib (+ best standard of care [BSC] [n=278]) and placebo (+ BSC [N=138]) arms with an 80% powering to detect a hazard ratio of 0.7 (corresponding to a median OS improvement from 6.3 to nine months). Inclusion criteria included a range of factors, but notably patients had to have been diagnosed with metastatic stage IV colorectal cancer, have failed on two prior treatments (with fluoropyrimidine, oxaliplatin and irinotecan) and were allowed to have prior anti-VEGF or anti-EGFR targeted therapy. Eastern Cooperative Oncology Group (ECOG) status, a measure of patient health, demonstrated that 72.3% (n=201) of fruquintinib patients had an ECOG status of 1 and 27.7% had a status of ECOG 0 (placebo; ECOG 1:73.2% [n=101], ECOG 0:26.8% [n=37]). ECOG 0 patients are asymptomatic and ECOG 1 patients are symptomatic and restricted in strenuous physical activity, but can perform the majority of normal day-to-day functions (ECOG scale goes up to 5, which is death).
Positive efficacy in difficult patient population
Both overall survival and progression-free survival (PFS) (Exhibits 1 and 2) demonstrated statistical significance over placebo, a notable clinical achievement when considering the difficult patient population in which fruquintinib was tested. Key points include:
■
Positive hazard ratio of 0.65 (95% CI: 0.51-0.83).
■
Median overall survival (OS) of 9.30 months in the fruquintinib group (95% CI 8.18-10.45) vs 6.57 months in the placebo group (95% CI 5.88-8.11) (p-value < 0.001).
■
Median progression-free survival (PFS) was 3.71 months (95% CI 3.65-4.63) vs 1.84 (95% CI 1.81-1.84) for placebo with a hazard ratio of 0.26 (95% CI 0.21-0.34) (p-value <0.001).
■
The majority of responses in the treatment arm were stabilisation of disease at 57.6% (n=160). Comparatively, 12.3% of placebo patients (n=17) had stabilised disease, with the majority (71%, n=98) having progressive disease.
■
One and 12 patients in the fruquintinib arm had a complete response and partial response respectively, while no patients experienced either in the placebo arm.
■
Disease control rate (CR+ PR + SD, > 8 weeks after randomisation, p<0.001) of 62.2% (n=173) vs 12.3% (n=17) was demonstrated.
These data compare favourably to both pivotal data from an international trial (CORRECT) with Bayer’s Stivarga (regorafenib) (a multi-kinase inhibitor targeting VEGFR 1,2 and 3, which is approved and sold in multiple countries) and to a subgroup analysis focused on Chinese patients from an Asian study (CONCUR). Both trials were in patients with previously treated metastatic colorectal cancer. Median overall survival on Stivarga in the CORRECT study was 6.4 months (95% CI, 5.8-7.3) versus 5.0 months (95% CI, 4.4-5.5) on placebo with a 0.77 hazard ratio (95% CI, 0.64-0.94). PFS was 2.0 months (95% CI, 1.9-2.3) vs 1.7 months on placebo (95% CI, 1.7-1.8).
Subgroup analysis of the CONCUR trial focused on Chinese patients (n=172, 112 on Stivarga, 60 on placebo). ECOG status of 1 was demonstrated in 72% of patients (n=81) and 0 in 28% of patients (31). OS in this Chinese patient subgroup was 8.4 months for Stivarga (vs 6.2 months on placebo, no CI given) with a hazard ratio of 0.56 (95% CI, 0.39-0.80). Median PFS was two months vs 1.7 on placebo (no CI given) with a hazard ratio of 0.32 (95% CI, 0.22-0.47).
Exhibit 1: FRESCO overall survival
|
|
Source: Hutchison China MediTech
|
Favourable efficacy and safety in comparison to competition
Safety will be a key differentiator for fruquintinib as it looks to position itself in the Chinese market as a best-in-class product. Grade 3 and above adverse events (AEs) occurred in 61.1% (n=170) of patients receiving fruquintinib versus 19.7% (n=27) of patients in the placebo arm. Dose interruptions, a key indicator of tolerability, were 35.3% (n=98) in the fruquintinib arm and 10.2% (n=14) in the placebo arm, while treatment discontinuation occurred 15.1% (n=42) and 5.8% (n=8) in each arm respectively. The most common grade 3-5 adverse events were hypertension (21.2%, n=59), palmar-plantar erythrodysaesthesia/hand-foot syndrome (HFSR/PPE) (10.8%, n=30) and proteinuria (3.2%, n=9). All other grade 3-5 AEs occurred at a frequency below 3%, further reinforcing the selective nature of fruquintinib.
Exhibit 2: Progression-free survival
|
|
Source: Hutchison China MediTech
|
Due to varying patient demographics and standards of care, trial comparisons, while useful, can be complicated. Dose interruptions can be a key indicator for tolerability and can negatively affect the anti-angiogenic mode of action of VEGFR inhibitors.
In the CORRECT trial testing Stivarga (regorafenib), it was reported that 61% of patients receiving Stivarga required a dose interruption. Similar results occurred in the CONCUR trial, which demonstrated that 68.8% of patients had a dose interruption. In the CORRECT and CONCUR trials, treatment discontinuation occurred in 8.2% and 14% of the patients respectively.
In the CORRECT trial the most common grade >3 AEs were HFSR/PPE (17% vs 0% placebo), fatigue (15% vs 9% placebo), pain (9% vs 7% placebo), infection (9% vs 6% placebo), hypertension (8% vs <1% placebo), diarrhoea (8% vs 2% placebo), rash (6% vs >1% placebo) and mucositis (4% vs 0% placebo). All other grade > 3 AEs occurred below 3%. Stivarga also contains a black box warning for fatal hepatotoxicity (liver damage), which occurred in 1.6% of the metastatic CRC patient population (vs 0.4% on placebo).
In the Chinese patient subgroup analysis of the CONCUR trial, grade 3,4 and 5 AEs occurred in 52% (n=58), 8% (n=9) and 10% (n=11) of patients on Stivarga and 35% (n=21), 2% (n=1) and 10% (n=6) on placebo, respectively.