The OBR Blog

January 22, 2016 - 08:01 pm Posted in ASCO GI Conference Coverage comments0 Comments

By: Arnold DuBell, Ph.D., M.B.A., Consultant, Clinical & Scientific Assessment, Kantar Health and Stephanie Hawthorne, Ph.D., Vice President, Clinical & Scientific Assessment, Kantar Health 

It seems impossible to attend an oncology conference these days without immuno-oncology being a significant focus, and it’s starting to become passé. Well, not quite. Two years ago the idea of this treatment class playing a role in gastrointestinal (GI) cancers seemed unlikely. However, as we learned yesterday at the 2016 American Society of Clinical Oncology (ASCO) Gastrointestinal (GI) Cancers Symposium, these tumors might be immune-sensitive after all. Updated data and new results were reported for several checkpoint inhibitors, which collectively support the continued development of these drugs in GI cancers.

In gastric cancer, first results of the CheckMate-032 trial were presented for Opdivo® (nivolumab, Bristol-Myers Squib/Ono Pharmaceuticals).1  In a heavily pretreated population of 59 patients with gastric or gastroesophageal junction (GEJ) cancer, single-agent Opdivo (at a dose of 3 mg/kg) resulted in a 14% objective response rate (ORR), which included one patient with a complete response. Among the responding patients, median duration of response was 7.1 months. Across all patients, the median progression-free survival (PFS) was 1.4 months, the median overall survival (OS) was 5.0 months, the six-month OS rate was 49%, and encouragingly 36% of patients were alive at one year. There was a slight correlation between ORR and PD-L1 expression, with higher ORR observed in PD-L1-positive patients compared with PD-L1–negative patients (27% versus 12%, using expression in 1% or more of the tumor cells as the definition of PD-L1-positivity). Consistent with studies in other tumor types, Opdivo was well tolerated, with only 17% of patients experiencing a Grade 3 or 4 adverse event. The efficacy data for Opdivo appear, at first glance, to be in line with the activity previously reported for Keytruda® (pembrolizumab, Merck & Co.): 22% ORR, 1.9-month median PFS, and 69% six-month OS in heavily pretreated PD-L1-positive gastric cancer.2

Also debuting data in gastric/GEJ cancer at ASCO GI 2016 was avelumab (MSB0010718C, Merck KGaA/Pfizer). The gastric cancer cohort of the JAVELIN Solid Tumor trial enrolled 75 patients and treated them with avelumab monotherapy in either the second-line setting (n=20) or as maintenance therapy following first-line chemotherapy (n=55).3  In second-line, avelumab produced a 15% ORR and 2.7-month PFS among all patients, and like the Opdivo results there was a suggested correlation between ORR and PD-L1 expression, with higher ORR observed in PD-L1-positive patients compared with PD-L1–negative patients (20% versus 0%, using 1% or more as the definition of PD-L1 positivity along with caution due to small sample sizes). Similar outcomes were reported in the JAVELIN Solid Tumor Japan trial’s gastric cancer cohort: 15% ORR in all patients and suggested correlation with PD-L1 expression (40% in PD-L1-positive versus 7% in PD-L1-negative patients, using 1% or more as the definition of PD-L1 positivity and caution due to small sample sizes).4

Although these study data are all preliminary, they serve collectively to provide evidence that checkpoint inhibition has clinical activity in metastatic gastric cancer and support the multiple Phase III trials that are already underway for all three of these drugs. While these data also suggest PD-L1 overexpression might correlate with improved ORR, the data are reminiscent of outcomes observed in other tumors and further perpetuate ideas that PD-L1 might not be the best biomarker to select patients for treatment with a checkpoint inhibitor.

Results were also reported for this class in metastatic esophageal cancer. The Ono Pharmaceuticals-sponsored Phase II ONO‐4538‐07 study enrolled 65 squamous cell carcinoma esophageal cancer patients and treated them with single-agent Opdivo (3 mg/kg q2w). In this heavily pretreated population, Opdivo produced a 17% ORR by central review, with 1.5-month median PFS and 10.8-month median OS.5  Among those patients who achieved a response, the duration is quite long (median was not reached, but all responding patients seemed to remain in response for at least eight months based on a review of the spider plot). This trial did not provide an analysis of outcomes according to patient PD-L1 status or other biomarker.

The KEYNOTE-028 trial evaluated Keytruda (10 mg/kg q2w) in relapsed/refractory metastatic esophageal cancer patients, specifically enrolling only PD-L1-positive patients, defined as expression in 1% or more of tumor or inflammatory cells or positive bands in the stroma. In  patients with predominantly squamous cell carcinoma treated in the third-line or later, the observed ORR of 30% among evaluable patients (n=23) was quite encouraging.6 These efficacy data were a reprisal of data originally presented at ASCO 2015,7 but of particular interest was new data presented on a gene expression signature composed of six genes that play a role in adaptive immune response. In tumors with a high inflamed signature score, there was a propensity toward higher likelihood of prolonged PFS compared with patients with a low score. Even among patients with a high signature score, there still appeared to be a subpopulation with greater benefit, so more work is needed to fully understand how to identify the most responsive patients with PD-1 inhibition, but this data was an intriguing step in the move beyond PD-L1 immunohistochemistry assays.

As in gastric cancer, development of PD-1 inhibitors is advancing quickly in esophageal cancer. Opdivo and Keytruda have recently initiated Phase III trials in this disease, with both targeting the relapsed/refractory setting.

The data for checkpoint inhibitors as reported at ASCO GI 2016 is focused within gastric, GEJ and esophageal cancers. However, previously reported data for Keytruda in colorectal cancer (CRC) was discussed, especially within the context of patient selection for use of checkpoint inhibitors. As reported originally at ASCO 2015, CRC and other GI tumors with high microsatellite instability (MSI-H) or that are deficient in mismatch repair (dMMR) have demonstrated high ORR and prolonged PFS when treated with Keytruda.8 Confirmation of this activity is ongoing in Phase II and Phase III studies. Clinical trials are also currently ongoing with Keytruda and Opdivo in pancreatic cancer and hepatocellular carcinoma.  Although we won’t see clinical data in these indications at ASCO GI 2016, today’s Keynote address by Dr. Robert Vonderheide presented intriguing in vitro and in vivo data that supports pursuit of immunotherapy in pancreatic cancer through a combination approach of immune response activation (using chemotherapy, radiotherapy, and/or vaccination) and immune response maintenance (using checkpoint inhibition).9 For now, checkpoint inhibitors continue to garner significant attention across a range of cancer indications, and gastrointestinal cancers can no longer be considered immuno-insensitive.


  1. Le DT, Bendell JC, Calvo E, et al.; “Safety and activity of nivolumab monotherapy in advanced and metastatic (A/M) gastric or gastroesophageal junction cancer (GC/GEC): Results from the CheckMate-032 study;” J Clin Oncol 34, 2016 (suppl 4S; abstr 6).
  2. Muro K, Bang Y-J, Shankaran V, et al.; “Relationship between PD-L1 expression and clinical outcomes in patients (Pts) with advanced gastric cancer treated with the anti-PD-1 monoclonal antibody pembrolizumab (Pembro; MK-3475) in KEYNOTE-012;” J Clin Oncol 33, 2015 (suppl 3; abstr 3).
  3. Chung HC, Arkenau HT, Wyrwicz L, et al.; “Safety, PD-L1 expression, and clinical activity of avelumab (MSB0010718C), an anti-PD-L1 antibody, in patients with advanced gastric or gastroesophageal junction cancer;” J Clin Oncol 34, 2016 (suppl 4S; abstr 167).
  4. Nashina T, Shitara K, Iwasa S, et al.; “Safety, PD-L1 expression, and clinical activity of avelumab (MSB0010718C), an anti-PD-L1 antibody, in Japanese patients with advanced gastric or gastroesophageal junction cancer;” J Clin Oncol 34, 2016 (suppl 4S; abstr 168).
  5. Kojima T, Hara H, Yamaguchi K, et al.; “Phase II study of nivolumab (ONO-4538/BMS-936558) in patients with esophageal cancer: Preliminary report of overall survival;” J Clin Oncol 34, 2016 (suppl 4S; abstr TPS175).
  6. Doi T, Piha-Paul SA, Jalal SI, et al.; “Updated results for the advanced esophageal carcinoma cohort of the phase Ib KEYNOTE-028 study of pembrolizumab (MK-3475);” J Clin Oncol 34, 2016 (suppl 4S; abstr 7).
  7.  Doi T, Piha-Paul SA, Jalal SI, et al.; “Pembrolizumab (MK-3475) for patients (pts) with advanced esophageal carcinoma: Preliminary results from KEYNOTE-028;” J Clin Oncol 33, 2015 (suppl; abstr 4010).
  8. Le DT, Uram JN, Wang H, et al.; “PD-1 blockade in tumors with mismatch repair deficiency;” J Clin Oncol 33, 2015 (suppl; abstr LBA100).
  9. Vonderheide RH, “Inflammatory networks and cancer immune surveillance;” Keynote address; American Society of Clinical Oncology Gastrointestinal Cancers Symposium; San Francisco, California; February 22, 2016.

January 21, 2015 - 10:01 am Posted in ASCO GI Conference Coverage comments0 Comments

Patients with metastatic colorectal cancer (mCRC) are typically treated at some point with at least one angiogenesis inhibitor. According to Kantar Health’s CancerMPact® Treatment Architecture U.S. data, Avastin® (bevacizumab, Genentech/Roche) is offered to approximately one-half of KRAS wild-type chemotherapy-naïve patients and three-quarters of KRAS mutated chemotherapy-naïve patients in the first-line setting. In the second- or third-line setting, patients might be treated with Avastin, Zaltrap® (ziv-aflibercept, Sanofi) or Stivarga® (regorafenib, Bayer/Onyx). Given data from the 2015 ASCO Gastrointestinal Cancers Symposium, Cyramza® (ramucirumab, Eli Lilly) may soon be added to the list of agents for second- or third-line therapy.

The use of Cyramza in this setting was examined in the Phase III RAISE trial, which enrolled 1,050 mCRC patients who had progressed during or after first-line Avastin, oxaliplatin and a fluoropyrimidine and randomized them to treatment with FOLFIRI q2w or FOLFIRI plus Cyramza 8 mg/kg q2w.1 Treatment would continue until disease progression or unacceptable toxicity. The trial roughly included similar numbers of KRAS wild-type and mutant patients, and this status was a stratification factor for the trial.

RAISE met its primary endpoint of improving overall survival (OS) as the addition of Cyramza reduced the risk of death by 16% (median OS: 13.3 months versus 11.7 months, HR 0.84, p=0.0219). Both KRAS wild-type and mutant patients achieved benefit, although neither subgroup analysis was statistically significant. RAISE also met a secondary endpoint of progression-free survival (median PFS: 5.7 months versus 4.5 months, HR 0.79, p=0.0005) but did not significantly improve the response rate (13.4% versus 12.5%, p=0.6336) or disease control rate (74.1% versus 68.8%, p=0.0587). The toxicity profile was not surprising given Cyramza’s current approvals in both relapsed/refractory gastric cancer and non-small cell lung cancer (NSCLC). Common Grade 3 or higher toxicities that were significantly increased with Cyramza included neutropenia (38.4% versus 23.3%), fatigue (11.5% versus 7.8%), hypertension (10.8% versus 2.8%) and thrombocytopenia (3.0% versus 0.8%).

Given the statistically significant improvement in overall survival, Cyramza should ultimately be approved by the U.S. Food and Drug Administration (FDA) for use in this setting. Although the degree of benefit in RAISE borders on clinically meaningful (1.6-month OS benefit, HR 0.84, p=0.0219), the FDA approved it in December for use in second-line NSCLC based on similar borderline benefit in the REVEL trial2 (1.4-month OS benefit, HR 0.857, p=0.0235). In both of these trials, the magnitude of benefit fell short of the Recommended Targets for Meaningful Clinical Trial Goals that were put forth by an ASCO working group in early 2014 for these two indications.3 Nevertheless, the FDA approved Cyramza in NSCLC and may do the same in mCRC given the demonstrated OS benefit.

However, earning regulatory approval may be the least of Cyramza’s difficulties for gaining future utilization in mCRC. Once approved, Cyramza will compete in the second-line setting with Avastin and Zaltrap as noted above. Both of these agents have been tested in patients pretreated with Avastin. Second-line Avastin in Avastin pretreated patients was directly examined in the Phase III ML 18147 trial, colloquially referred as TML (for Treatment across Multiple Lines).4 TML evaluated the addition of Avastin to chemotherapy (either oxaliplatin- or irinotecan-based). Zaltrap was approved for second-line use based on data from the Phase III VELOUR study, which evaluated the addition of Zaltrap to FOLFIRI in second-line patients. Subgroup analysis in patients who were treated with first-line Avastin has been published.5 As shown in the table below, all three anti-angiogenic agents show similar underwhelming (>0.80) hazard ratios for OS.

Efficacy Comparisons Between Avastin, Zaltrap and Cyramza in 2nd-Line mCRC
  TML (Avastin-Pretreated3) VELOUR (Prior Avastin Subgroup Only4) RAISE (Avastin-Pretreated1)
  Placebo + CT (n=410) Avastin + CT (n=409) HR (p-value) Placebo + FOLFIRI (=187) Zaltrap + FOLFIRI (n=186) HR (p-value) Placebo + FOLFIRI (n=525) Cyramza + FOLFIRI (n=525) HR (p-value)
Median OS 9.8 11.2 0.81 (p=0.0211) 11.7 12.5 0.862 (NP) 11.7 13.3 0.84 (p=0.0219)
Median PFS 4.1 5.7 0.68 (p<0.0001) 3.9 6.7 0.661 (NP) 4.5 5.7 0.79 (p=0.0005)
CT: Chemotherapy; NP: Not provided


Not only is Cyramza attempting to compete with agents that have been approved for this setting since 2012 (Zaltrap) and 2013 (Avastin for use in Avastin-pretreated CRC), but it may be doing so with a pricing problem. One physician in attendance at the ASCO GI presentation directly asked why he should offer Cyramza if it costs “twice as much” as Avastin. Estimated costs of these agents support this claim: Avastin per cycle, assuming the use of the lower dosage of 5 mg/kg q2w, is approximately $2,400. Cyramza is expected to cost $5,900 per cycle of therapy based on the dose used in this trial and its current price for approved indications. This is eerily reminiscent of difficulties for Zaltrap, when several physicians from the Memorial Sloan-Kettering Cancer Center wrote an opinion piece in The New York Times stating they would not provide Zaltrap to their patients since it was more than twice the cost of Avastin; the negative publicity associated with the article forced Sanofi to drop its pricing for Zaltrap.  

The situation doesn’t improve for Cyramza even if it is considered in later lines of therapy. Not only would Cyramza need to compete with the anti-angiogenic agent Stivarga, which was approved in 2012 by the FDA, but it also will likely need to compete against TAS-102 (Lonsurf® in Japan, Taiho Oncology), a formulation of two antimetabolic agents that met its primary endpoint of OS in the Phase III RECOURSE study.6 The FDA has granted TAS-102 a fast-track designation, and Taiho confirmed that it had already initiated a rolling submission for approval.

Cyramza faces tough hurdles ahead given the data presented at the ASCO Gastrointestinal Cancers Symposium: activity comparable to other anti-angiogenic agents that are already approved in the same setting, and at a potentially higher cost. With these hurdles, it is hard to imagine Cyramza “RAISEing” the bar in second-line CRC in the near future.


1. Tabernero et al.; Abstract 512, 2015 ASCO Gastrointestinal Cancers Symposium
2. Perol, et al.; Abstract LBA8006, ASCO 2014
3. Ellis, et al.; J Clin Oncol, 32(12): 1277-1280, 2014.
4. Bennouna, et al.; Lancet Oncol, 2013
5. Tabernero, et al.; Eur J Cancer, 2014
6. Yoshino, et al.; Abstract O-0022, 2014 ESMO World Congress on Gastrointestinal Cancer.

By: Stephanie Hawthorne, Ph.D., Senior Director, Clinical and Scientific Assessment, Kantar Health and Arnold DuBell, Ph.D., M.B.A, Consultant, Clinical and Scientific Assessment, Kantar Health

January 16, 2015 - 09:01 am Posted in ASCO GI Conference Coverage comments0 Comments

Designated as a “breakthrough therapy” by the FDA for both advanced and unresectable malignant melanoma and advanced non-small cell lung cancer (NSCLC), and approved by the U.S. Food and Drug Administration (FDA) in September 2014 for advanced pretreated metastatic melanoma, the immune checkpoint inhibitor Keytruda® (pembrolizumab, Merck) has also shown promise for several other indications including (but not limited to) head and neck cancer, Hodgkin’s lymphoma and triple-negative breast cancer (TNBC).

Given Keytruda’s remarkable performance as an antitumor agent in several advanced/metastatic diseases, the therapeutic potential of this promising new drug is now being explored in the context of advanced gastric cancer. In a keynote address on Thursday, Jan. 15 at the 2015 American Society of Clinical Oncology (ASCO) Gastrointestinal Cancers Symposium, an updated cohort analysis of 39 patients with either recurrent or metastatic adenocarcinoma of the stomach or gastroesophageal junction from the Phase 1b KEYNOTE-012 study1 was presented. For this specific analysis, only patients with distinctive stromal or ≥1% tumor nest cell PD-L1 staining were included. Patients were heavily pretreated; 67% of the patients had received two or more prior therapies. Patients were given 10 mg/kg Keytruda every two weeks for up to 24 months or until complete response, progression or unacceptable toxicity. The primary efficacy endpoint was objective response rate (ORR), and secondary endpoints were duration of response, progression-free survival (PFS) and overall survival (OS).

Keytruda achieved a favorable ORR of 22% by central review and 33% by investigator review. This differs only slightly from that presented at the European Society for Medical Oncology (ESMO) 2014 Congress (30.8% ORR by investigator review).2 Responses were similar between Asian and non-Asian patients. The median time to response was eight weeks, and the median response duration was 24 weeks. The six-month PFS rate was 24%, and the median PFS was 1.9 months. The six-month OS rate was an impressive 69%, and the median OS was not yet reached. Patients had a median follow-up duration of 8.8 months, and 33% remained on therapy as of November 2014. Although the data were very preliminary, a trend toward an association between higher levels of PD-L1 expression and ORR (p=0.102), PFS (p=0.162) and OS (p=0.124) was observed. Only four patients experienced Grade 3 or higher drug-related adverse events: peripheral sensory neuropathy, fatigue, decreased appetite, hypoxia and pneumonitis (2.6% each); one of these adverse events (hypoxia) resulted in patient death.

To date, the treatment landscape for advanced or relapsed gastric cancer has proven to be relatively bleak. Prior to the approval of the VEGFR-2-targeted therapy Cyramza® (ramucirumab, Eli Lilly), there were no approved targeted agents for the treatment of relapsed gastric cancer, and physicians relied on fluoropyrimidine- or taxane-based chemotherapy. Cyramza was approved in the U.S. and Europe based on results from the REGARD and RAINBOW trials. In REGARD, Cyramza was associated with an OS of 5.2 months versus 3.8 months for placebo (HR 0.776, p=0.0473) in patients previously treated with first-line platinum- or fluoropyrimidine-based therapy. In addition, the six-month OS rates were 41.8% in patients treated with Cyramza versus 31.6% in the placebo group.3 RAINBOW compared paclitaxel ± Cyramza as a second-line therapy; the addition of Cyramza to paclitaxel significantly extended median OS (9.6 months versus 7.4 months, HR 0.81, p=0.017). The six-month and 12-month OS rates for Cyramza plus paclitaxel were 72% and 40%, respectively, compared with 57% and 30% for paclitaxel alone.4

The results of the KEYNOTE-012 gastric cancer cohort analysis highlight the promising antitumor activity and manageable toxicity of Keytruda in advanced gastric cancer. While there are caveats to comparing trials directly, patients treated with Keytruda monotherapy had a similar six-month OS (69%), as did patients treated with Cyramza + paclitaxel (72%); however, recall that the patients on Keytruda were more heavily pretreated. Building on this signal, Merck has announced plans to initiate a Phase II trial (KEYNOTE-059; NCT02335411) to evaluate Keytruda or Keytruda plus cisplatin and 5-FU in 270 patients with recurrent or metastatic gastric and gastroesophageal junction adenocarcinoma. The primary outcomes will be safety and response. This study will provide some important insights on the activity and safety of Keytruda, including its use in combination with chemotherapy, the importance of PD-L1 as a biomarker, and the role of Keytruda in both first-line and relapsed settings. Findings from KEYNOTE-012 should provide valuable to inform future Phase III trial design.

However, as with other tumor types, other manufacturers of PD-1 inhibitors have announced plans to compete in this space. A new Phase III trial (NCT02267343) was initiated in October 2014 to compare Opdivo® (nivolumab, ONO-4538/BMS-936558, Bristol-Myers Squibb/Ono Pharmaceuticals) versus placebo in 480 previously treated Japanese patients with unresectable advanced or recurrent gastric and gastroesophageal junction cancer. This trial will not limit patient enrollment by PD-L1 biomarker status. The primary endpoint will be OS, and secondary endpoints will include PFS, ORR and safety. To date, there are no known late-stage trials of Opdivo in Western gastric cancer patients, but it can be assumed that trials may be initiated in the near future. As in melanoma, Opdivo may ultimately have a first-to-market advantage over Keytruda in gastric cancer in Japan, but Keytruda may have a leg up in the Western markets for this indication.

Along with immunotherapy, other targeted therapies are planning Phase II trials, including Stivarga® (regorafenib, Bayer, NCT01913639) and AMG 337 (Amgen, NCT02016534). Eli Lilly has also announced plans for another Phase III trial for Cyramza (RAINFALL, NCT02314117), which will compare capecitabine plus cisplatin with or without Cyramza in 616 newly diagnosed gastric or gastroesophageal junction cancer patients.

As with other tumors, the competitive landscape for gastric could dramatically change within the next several years given the data presented. These data are certainly at an early stage, but the high response rates and preliminary OS data justify the excitement associated with the PD-1 inhibitors and add gastric cancer to the growing list of tumors in which immuno-oncology may play a significant role.


  1. Muro et al., Abstract 3, ASCO GI 2015.
  2. Muro et al, Abstract LBA15, ESMO 2014.
  3. Fuchs et al., Lancet. 2014 Jan 4;383(9911):31-9.
  4. Wilke et al., Lancet Oncol. 2014 Oct;15(11):1224-35.

By: Stephanie Hawthorne, Ph.D., Senior Director, Clinical and Scientific Assessment, Kantar Health and Elizabeth Clarke, Ph.D., Analyst, Clinical and Scientific Assessment, Kantar Health

January 21, 2014 - 09:01 am Posted in ASCO GI Conference Coverage comments0 Comments

Clinically speaking, hepatocellular carcinoma (HCC) is a particularly challenging disease. Surgery remains the only curative strategy for patients with HCC, however not all patients are eligible for surgery. The approval of Nexavar® (sorafenib, Bayer/Onyx/Amgen) revolutionized HCC, providing a targeted therapy option for patients with HCC and an advance over de facto standard doxorubicin. The enthusiasm for Nexavar is dampened by the fact that not all patients may be eligible for Nexavar treatment due to poor liver function, and options remain limited for patients whose disease progresses. Since the introduction of Nexavar, other targeted therapies have attempted to carve a place in the HCC treatment landscape, either as frontline agents or in the relapsed setting, only to be met with failure.

Recent pivotal trial failures include: Sutent® (sunitinib, Pfizer) in comparison with Nexavar in first-line;1 brivanib compared with Nexavar in first-line2 and in Nexavar-refractory patients;3 and most recently Afinitor® (everolimus, Novartis) failed to demonstrate efficacy in Nexavar-refractory HCC patients.4 Despite multiple failures, there is a long list of agents that are currently in active clinical development either in Phase II or Phase III trials, with most agents aiming for the Nexavar-refractory population. Agents currently in pivotal trials in Nexavar-refractory patients include ADI-PEG20 (Polaris Group), Cometriq® (cabozantinib, Exelixis), ramucirumab (Eli Lilly), and Stivarga® (regorafenib, Onyx/Amgen), as well as tivantinib (Arqule/Daiichi Sankyo) that is being studied in a subpopulation of patients with MET overexpression.

Into this matrix enters LY2157299, a novel transforming growth factor-beta receptor 1 (TGF-β1) kinase inhibitor that Eli Lily is developing in HCC. Elevated TGF-β1signaling promotes liver fibrosis and progression to HCC; hence targeting TGF-β signaling has been proposed as a unique approach in HCC treatment. While limited data is available with LY2157299, this agent prevented HCC cancer cell migration on extracellular matrix substrates in cell culture studies.5 Given the potential importance of this pathway in HCC and some preliminary evidence that inhibition of this pathway can inhibit certain processes in HCC, Eli Lilly is evaluating this compound in an ongoing Phase II study. The trial is evaluating LY2157299 in patients ineligible for Nexavar therapy or who have progressed on Nexavar. While the trial is still recruiting, interim results were presented6 at the American Society of Clinical Oncology (ASCO) Gastrointestinal Cancers Symposium on Friday, January 17, 2014. The study accrued 109 patients who were randomized to two arms: Arm A LY2157299 at 160 mg/day and Arm B LY2157299 at 300 mg/day. Primary endpoints were time-to-progression (TTP) and biomarker changes (serum AFP, TGF-β, and e-cadherin). Serum AFP was used as a surrogate marker for clinical efficacy and patients in whom AFP levels decreased greater than 20% compared to baseline were considered “AFP responders.” Of the eligible patients accrued, about 80% were Nexavar-pretreated and 50% had AFP levels greater than 400 ng/ml at baseline. The median TTP was 12.0 weeks in Nexavar-refractory patients (90% CI: 6.6-12.6) and 18.3 weeks in Nexavar-naïve patients (90% CI: 6.6-42.4). About one-quarter of the patients achieved more than a 20% decrease in AFP levels and were categorized as AFP-responders. The drug was very well-tolerated, with the most common grade 3/4 toxicities being neutropenia (2.7%) and fatigue (1.8%). Perhaps the most exciting piece of data from this study was the increase in median OS in the AFP responder population compared to AFP-non-responders (93.1 weeks versus 29.6 weeks; p=0.00006). In contrast, median OS in the entire population was 8.3 mos (36 weeks). Given that Nexavar achieves an OS of about 10 months and given that not all patients can tolerate Nexavar, this appears to be a niche positioning opportunity for Eli Lilly.

Although this study is very promising, there are a couple of caveats. Key amongst them is the use of AFP as a biomarker. Decreases in AFP levels have been shown to be clinically useful in monitoring tumor response in HCC patients; however, AFP detection is also known to be a poorly reliable marker, particularly in small tumors.7 This may present a challenge in metastatic disease or on small tumors that progress very slowly. Whether AFP is the most appropriate biomarker to use in this context and whether this is an appropriate biomarker to support approval down the line is not known – all biomarkers approved to-date focus on patient selection based on baseline presence of the biomarker; this would pave new ground as a biomarker of response post-therapy, and perhaps would be more useful in guiding decisions of whether to continue treating patients with the drug as opposed to guiding decisions of whether to initiate therapy with the drug. In addition, whether the inhibitor is truly inhibiting the TGF-βpathway is not known, as downstream targets such as phosphor-Smads have not yet been evaluated. The phosphorylation of Smad protein plays an important role in TGF-β signaling and activates a nuclear transduction protein.

So where does LY2157299 stand in terms of novel agents in HCC? Clearly, it is too early to determine how this compound will fit into the overall picture of HCC therapy. The table below provides a preview of the clinical efficacy of LY2157299 in comparison to the key agents in development in Nexavar-refractory patients.

Table 1: Efficacy Outcomes in Nexavar-pretreated HCC

In favor of LY2157299 are its manageable toxicity profile and the fact that it targets a novel pathway. However, the demonstrated efficacy benefit in the general population is not striking compared to the other agents currently in development in the same population. The OS and TTP observed with LY2157299 in AFP responders are particularly intriguing; however, whether the test is appropriate or whether AFP level is an appropriate marker for response remains a concern. While the study is promising, we will have to wait until the data mature and see how this plays out in a larger Phase III trial.


  1. Cheng et. al., J Clin Oncol,. 31(32): 4067-4075, 2013
  2. Johnson et al., J Clin Oncol, 31(28): 3517-3524, 2013.
  3. Llovet et. al., J Clin Oncol,31(28):3509-16, 2013.
  4. Zhu et al.. Abstract 172, ASCO GI 2014.
  5. Duturi et. al. PLOS One 27;8(6):e67109 2013
  6. Faivre, Abstract LBA 173, ASCO GI 2014
  7. Farinati et. al., Am J Gastroenterol, 101(3):524-32, 2006
  8. Bruix et al., Eur J Canc, 49(16): 3412-3419, 2013
  9. Verslype, Abstract 4007, ASCO 2012

By Neesha Suvarna, PhD, Consultant, Kantar Health and Len Kusdra, Analyst, Kantar Health

article register

Recent Posts

Recent Comments