By: Len Kusdra, Analyst, Clinical & Scientific Assessment, Kantar Health and Stephanie Hawthorne, Senior Director, Clinical & Scientific Assessment, Kantar Health
The introduction of Herceptin® (trastuzumab, Roche/Genentech) over a decade ago turned HER2+ metastatic breast cancer (MBC) from what was once a patient population with poor prognosis to one with a vastly improved outlook and led the way for an influx of targeted therapies for this population. Now, the armamentarium against HER2+ MBC has expanded and includes several HER2-targeted agents: Tykerb® (lapatinib, GSK), Perjeta® (pertuzumab, Roche/Genentech) and Kadcyla® (ado-trastuzumab, Roche/Genentech). Updated overall survival (OS) data from the CLEOPATRA (NCT00567190) trial were presented at the Presidential Symposium at ESMO on September 28, 20141, and the results will serve to further solidify Roche as the dominant player in HER2-targeted treatment in MBC, if there was any doubt before now. The CLEOPATRA trial was designed to test whether the combination of Herceptin, Perjeta and docetaxel improved outcomes in first-line HER2+ MBC patients. The study randomized 808 patients to receive Hercpetin plus docetaxel plus either placebo or Perjeta. The primary endpoint was progression-free survival (PFS), and secondary endpoints included OS, PFS as assessed by the investigator, overall response rate and safety.
The initially published results2 in 2012 showed an improvement in PFS from 12.4 months in the control group to 18.5 months in the Perjeta group (HR=0.62; 95% confidence interval (CI), 0.51-0.75; P<0.001). The objective response rate was 69.3% in the control group and 80.2% in the Perjeta group (95% CI, 4.2-17.5; P = 0.001). These significant positive results led to FDA approval in 2012 and European approval in 2013 for Perjeta in combination with Herceptin and docetaxel as first-line treatment in HER2+ MBC. At the time, however, the data was not yet mature to determine OS, and many were left wondering whether the impressive improvement in the primary endpoint of PFS would translate into an equally significant improvement in OS.
Final results of CLEOPATRA did not disappoint. With a median follow-up of 50 months (range 0-70 months), addition of Perjeta to Herceptin plus docetaxel provided a 15.7-month improvement in OS compared with patients who received Herceptin, docetaxel and placebo (mOS: 56.5 months in the Perjeta arm vs. 40.8 months in the placebo arm; HR=0.68, 95% CI 0.56-0.84, p=0.0002). Updated PFS data were similar to what was published in 2012, with a slight increase in PFS in the Perjeta arm (18.7 months vs. 12.4 months, HR=0.68, p<0.0001). The toxicity profiles between the arms were similar and manageable, with the Perjeta arm experiencing a higher rate of Grade 3 febrile neutropenia (13.7% vs. 7.6%) and diarrhea (9.3% vs. 5.1%). The significant improvement in OS now solidifies Perjeta plus Herceptin and docetaxel as the new standard care in first-line MBC patients.
So what is next? According to Kantar Health’s CancerMPact® Treatment Architecture, Perjeta has already benefited from a significant penetration into the first line setting, with use in one-third of U.S. patients in the year after its launch.3 The question that arises is, can we improve even further on these results, and can we do it by eliminating chemotherapy altogether? One intriguing piece of data presented was the duration of study treatment: While the number of cycles of docetaxel remained the same at eight cycles (range: 1-42 cycles for the placebo arm and 1-52 cycles for the Perjeta arm), the duration of study treatment was extended with the addition of Perjeta from 11.4 months (range: 0.1-66.3 months) in the placebo arm to 17.4 months (range: 0.1-67.7 months) in the Perjeta arm, raising the possibility of prolonged Perjeta and Herceptin combination even after ending the docetaxel portion of the study treatment. The possibility of removing chemotherapy and replacing it with targeted therapy alone has been the Holy Grail in the oncology field ever since Herceptin revolutionized the way MBC is treated. In that vein, the MARIANNE (NCT01120184) study is looking to do just that by examining whether Perjeta and Kadcyla are more efficacious than Herceptin plus Perjeta and a taxane (paclitaxel or docetaxel). In essence, Kadcyla would provide both the targeted therapy and the microtubule-disrupting activity provided by taxanes. Results from this study would certainly provide further impetus for a paradigm shift of complete elimination of chemotherapy in first-line MBC. Results from MARIANNE are expected before the end of this year. Until then, the results from the CLEOPATRA trial provide great hope for patients in a disease that had few options over a decade ago and opens the door for the development of novel and more potent combination therapies in what was once an intractable disease. These are exciting times indeed.
By: Mara Jeffress, Associate Consultant, Clinical & Scientific Assessment, Kantar Health and Stephanie Hawthorne, Senior Director, Clinical & Scientific Assessment, Kantar Health
Head and neck cancer patients tend to receive chemotherapy doublets with or without the only approved targeted agent, the EGFR antibody Erbitux® (cetuximab, Bristol-Myers Squibb/Lilly/Merck Serono) as first-line therapy. Up to 90% of squamous cell carcinoma of the head and neck (SCCHN) patients have a tumor with EGFR aberrations, and according to Kantar Health’s CancerMPact® Treatment Architecture United States, in 2013 38% of chemotherapy-naïve SCCHN patients received Erbitux as part of their first-line therapy for advanced disease. Erbitux may also be used in second-line, with approximately one-fifth of patients receiving Erbitux monotherapy and another one-fifth receiving Erbitux in combination with chemotherapy.
The use of Erbitux in multiple lines belies the large unmet need for relapsed or refractory patients, whose median overall survival ranges from three to six months (Machiels, Abstract LBA29, ESMO 2014). In addition, less than half of patients who receive a first-line therapy go on to receive a second line of therapy.1 Several other EGFR-directed agents have been tested in SCCHN, including Vectibix® (panitumumab, Amgen), zalutumumab (Genmab), Iressa® (gefitinib, AstraZeneca), and Tykerb® (lapatinib, GlaxoSmithKline). All have failed to improve survival in Phase III trials. Gilotrif® (Giotrif® in Europe, afatinib, Boehringer Ingelheim), an oral small molecule that inhibits EGFR, HER2 and HER4, is currently approved in non-small cell lung cancer (NSCLC) and is vying for a spot in SCCHN.
The LUX-Head & Neck 1 trial of Gilotrif in 474 second-line SCCHN patients who had progressed on platinum treatment advanced the field at this year’s European Society for Medical Oncology (ESMO) meeting2 by being the first positive trial of a novel targeted agent since the pivotal Erbitux EXTREME trial.3 The LUX-Head & Neck 1 trial stratified patients 2:1 to receive either Gilotrif (40 mg daily, oral) or methotrexate (40 mg/m2, IV weekly), and patients were stratified by ECOG status (0 vs. 1) and prior EGFR inhibitor use. Median progression-free survival (PFS) was significantly improved in the Gilotrif arm (2.6 vs. 1.7 months, HR=0.80, p=0.030), as was overall response rate (ORR; 10.2 vs. 5.6%) and disease control rate (DCR; 49.1 vs. 38.5%). There was no significant difference in overall survival (median OS 6.8 vs. 6.0 months, HR=0.96, p=0.7). Patient-reported outcomes and quality-of-life measures favored Gilotrif. In addition, Gilotrif was less toxic with fewer dose reductions (32% vs. 42%), discontinuations (7% vs. 16%) and fatal events (0.6% vs. 3%). As expected, adverse events characteristic of EGFR inhibitors were higher in the Gilotrif arm (all grade rash 74% vs. 8%; Grade 3/4 rash 10% vs. 0%; all grade diarrhea 72% vs. 12%; Grade 3/4 diarrhea 10% vs. 2%). In the methotrexate arm stomatitis (39% vs. 43%), fatigue (25% vs. 32%) and neutropenia (<1% vs. 19%) were higher.
Subgroup analysis from LUX-Head & Neck 1 suggests that patients who are HPV p16-negative and who have not received prior Erbitux are more likely to respond to Gilotrif, and prior Erbitux in 60% of the enrolled patients may have negatively biased the trial. How much better would the efficacy outcomes have been if patients had been preselected to be HPV-negative? With such a high number of patients being treated with Erbitux in front-line, would it have been possible to quickly enroll such a trial? Perhaps, as was suggested by the discussant, Dr. Seiwert, there are other biomarkers that might help refine which patients benefit the most from Gilotrif?
The suggestion that prior treatment with Erbitux negatively biased overall survival outcomes in this trial may be a biologically sound rationale, but it is clinically irrelevant since Erbitux is an approved drug used in a large proportion of patients with advanced disease. This outcome may speak to acquired resistance to EGFR inhibition, which is an inherent risk to developing a drug with a similar mechanism of action in the relapsed/refractory setting. That PFS was significantly improved despite the majority of patients having received prior Erbitux suggests they were not completely resistant, that Gilotrif remains active in Erbitux-resistant clones, or that the mechanisms of action of these two drugs (monoclonal antibody vs. tyrosine kinase inhibitor) differ sufficiently to confer response when used in sequence.
Similar outcomes have been observed with other EGFR inhibitors studied in the platinum-pretreated advanced SCCHN setting: Zalutumumab significantly improved PFS but not overall survival compared with best supportive care,4 and Iressa improved response rate but not PFS or overall survival when compared with methotrexate.5 The lack of overall survival benefit in LUX-Head & Neck 1 is concerning since there are so few effective options in this disease; one would expect that an active agent could have an impact on overall survival. The FDA may take a similar stance, requiring an overall survival benefit for approval, although other regulatory authorities (such as the European Medicines Agency) may be more willing to approve a drug with a PFS benefit alone.
If approved, Gilotrif will compete directly with Erbitux in platinum-pretreated patients in the U.S., although ex-U.S. it will be largely unchallenged due to lack of approval of Erbitux in second-line. In the U.S., Erbitux will have nearly a decade head start over Gilotrif, although Gilotrif could have the advantage of being the first marketed drug to show a PFS benefit in a randomized trial in this setting (Erbitux was approved in the U.S. based on a single-arm study). A comparison of the available data suggests that Gilotrif has similar efficacy in second-line as Erbitux; in 103 platinum-resistant second-line patients, Erbitux demonstrated a 2.3-month time to progression and a 12.6% response rate.6 With physicians being so familiar with Erbitux, would they be willing to switch to Gilotrif given the available data? The fact that Gilotrif is an oral agent will help it differentiate itself. However, given the higher co-pays required for oral agents in the United States, patients’ pocketbooks may disagree. At this time, no head-to-head trials of Erbitux vs. Gilotrif are planned.
The LUX-Head & Neck 1 trial results are the first to report as part of the larger LUX-Head & Neck program, which includes a companion trial of similar design conducted in Asian patients (LUX-Head & Neck 3), as well as two studies evaluating Gilotrif as post-remission therapy in locally advanced SCCHN (LUX-Head & Neck 2 and 4). It’s encouraging that PFS was improved in this first trial to report, and confirmation of activity in these other trials could go a long way to establish the degree of clinical benefit that Gilotrif confers in SCCHN and form a foundation upon which it may become adopted into clinical practice for this disease.
In recent years, angiogenesis inhibition, and VEGFR inhibition in particular, has been the most promising therapeutic target in advanced ovarian cancer, and the VEGF-targeted monoclonal antibody Avastin® (bevacizumab, Roche/Genentech) has been the major player in drug development for this tumor type. With Phase III trials conducted in multiple settings in ovarian cancer, Avastin has demonstrated a progression-free survival (PFS) benefit in first-line1,2,3, recurrent platinum-sensitive4, and recurrent platinum-resistant5,6 patients, and in the past two years, it has received regulatory approval in the European Union for use in the first-line/maintenance and recurrent platinum-sensitive settings. Other small molecule VEGFR inhibitors have followed Avastin into late stage development in ovarian cancer; Votrient® (pazopanib, GlaxoSmithKline) and cediranib (AstraZeneca, development discontinued) have demonstrated positive results in Phase III trials in the first-line maintenance and recurrent platinum-sensitive settings, respectively7, and an ongoing Phase III trial is evaluating Vargatef™ (nintedanib, Boehringer Ingelheim) as first-line/maintenance therapy.
Trebananib (Amgen) is a first-in-class agent that inhibits angiogenesis through a different pathway: the angiopoietin axis. Trebananib, a recombinant peptide-Fc fusion protein (peptibody), prevents the interaction of ligands angiopoietin 1 and 2 (Ang1 and Ang2) with the Tie2 receptor, thereby inhibiting a signaling pathway that is normally involved in vascular growth, remodeling, and stabilization. Amgen has initiated a comprehensive Phase III development program for trebananib in ovarian cancer, with the TRINOVA-1 and TRINOVA-2 trials in recurrent patients with platinum-resistant or partially platinum-sensitive disease and the TRINOVA-3 trial in the first-line/maintenance setting.
The results of the TRINOVA-1 trial were presented on Tuesday, October 1 at the 2013 European Cancer Congress (ECC)8. This double-blind Phase III trial (NCT01204749) randomized 919 recurrent patients with advanced ovarian, primary peritoneal, or fallopian tube cancer to receive weekly paclitaxel (80 mg/mg2) plus placebo or trebananib (15 mg/kg). Patients could have received up to 3 prior regimens and must have had a platinum-free interval (PFI) of less than 12 months after their most recent line of therapy; about half of patients in each arm had a PFI of less than 6 months, a threshold which is often used to define platinum resistance. The primary endpoint of the trial was PFS, and secondary endpoints included OS, ORR (using RECIST criteria), CA-125 response rate, and quality of life (FACT-O, OCS, and EQ-5D measurements).
The trial met its primary endpoint, with the addition of trebananib to paclitaxel significantly increasing PFS (see Table 1, mPFS: 7.2 vs. 5.4 months, HR: 0.66, p<0.001), and this benefit was consistent across multiple subgroups, including patients with a PFI less than 6 months as well as those with a PFI between 6 and 12 months. Patients with prior anti-angiogenic therapy (e.g. Avastin) were allowed in the study, but due to the large degree of variability in the PFS results for those patients (HR: 0.69, with 95% CI of 0.41 – 1.17), the efficacy results in that subgroup are inconclusive based on the data presented. In addition to PFS, the overall response rate was also improved in patients who received trebananib plus paclitaxel (38.4% vs. 29.8%, p<0.0071). However, although there was a slight numerical improvement, an interim analysis did not find a statistically survival benefit of adding trebananib to paclitaxel (mOS: 19.0 vs. 17.3 months, HR: 0.86, p=0.19); mature survival data are expected in 2014. Interestingly, the dose of trebananib used in TRINOVA-1 (15 mg/kg) was increased from the highest Phase II dose (10 mg/kg) in the hopes of improving drug efficacy. An analysis of the Phase II data found a trend toward an increased PFS in patients with a higher drug exposure (AUC of at least 9.6 mg*h/mL) that was not routinely reached in patients who received the 10 mg/kg dose9.
The toxicity profile of the trebananib and paclitaxel combination was manageable, and the addition of trebananib barely increased the incidence of Grade 3+ adverse events (56%) compared to paclitaxel alone (54%). Adverse events (of all grades) that were increased in the paclitaxel plus trebananib arm included localized edema (57% vs. 26%), peripheral neuropathy (21% vs. 16%, thought by the presenter to be related to increased paclitaxel exposure), ascites (20% vs. 12%), pleural effusion (13% vs. 4%), upper abdominal pain (12% vs. 7%), and generalized edema (11% vs. 3%). There was little to no increase in the incidence of typical VEGFR inhibition-related adverse events, such as hypertension (6% vs. 4%), proteinuria (3% in both arms), arterial thrombotic events (<1% in both arms), impaired wound healing (<1% in both arms), and GI perforation/fistula (1% vs. <1%), in the paclitaxel plus trebananib arm compared to pacliltaxel alone.
Considering these data, what will be trebananib’s regulatory and commercial fate? This question is best viewed through the prism of Avastin’s experience. To date, all Phase III trials evaluating Avastin in ovarian cancer have shown a significant PFS benefit, but no OS benefit. Based on these data, Avastin has been approved by the European Medicines Agency (EMA) for use in the first-line/maintenance setting as well as the recurrent platinum-sensitive setting; however, Roche has not yet filed for approved in Europe in the recurrent platinum-resistant setting. In contrast, Roche has not filed for approval in the U.S. in any setting in ovarian cancer, strongly suggesting that the FDA has provided guidance that a PFS benefit without an OS benefit would not be approvable. If trebananib’s experience with the European and U.S. regulatory agencies reflects that of Avastin, then it will likely be approved by the EMA with only the current PFS benefit, but if the mature OS data do not show a significant survival benefit, then it is unlikely to be approved by the U.S. Federal Drug Administration (FDA). However, it is possible that the high unmet need in platinum-resistant patients and the reasonable toxicity profile of trebananib might sway U.S. regulators in favor of the drug.
If approved, how will trebananib fare against Avastin, which will be its primary competitor in the recurrent setting? Based on the fact that the TRINOVA-1 trial include recurrent patients who were platinum-resistant as well as those who were partially platinum-sensitive, the patient population in this trial is not directly comparable to either the AURELIA or OCEANS trials of Avastin in recurrent platinum-resistant and platinum-sensitive patients, respectively. In general, however, the 1.8 month PFS benefit seen in the TRINOVA trial is somewhat smaller than the 3.3 month and 4.0 month PFS benefits observed in the AURELIA and OCEANS trials, respectively (see Table 2). Therefore, it is unlikely that physicians would routinely prefer trebananib over Avastin.
Due to their different mechanisms of action, however, Avastin and trebananib do have very distinct safety profiles, and this distinction could easily lead to the two agents being used in slightly different patient populations. Avastin administration generally results in typical VEGFR-related toxicities such as hypertension, proteinuria, and in rare cases, GI perforation and bleeding. In contrast, trebananib does not cause these VEGFR-related side effects, but does increase the incidence of edema, ascites, and pleural effusions. These side effect profiles, combined with recent data presented at the 2013 ECC suggest that first-line Avastin is more effective in higher-risk patients3, and also suggest that Avastin will be more likely to be utilized in high risk or highly symptomatic patients who already have baseline ascites and other similar symptoms. In contrast, trebananib might be more likely to be used in patients with co-morbidities that increase the risk of serious side effects from Avastin, such as a history of cardiovascular disease.
The data from the TRINOVA-1 trial also raise several additional questions about the use of these two agents. For example, how should these agents be sequenced? In patients who receive Avastin in the first-line setting, re-treatment with Avastin is unlikely, and that may be the primary opportunity for trebananib. Considering the non-overlapping safety profiles, another major outstanding question is whether trebananib and Avastin (or other VEGFR-targeted agents) should be used in combination. Although these questions will not be answered conclusively anytime soon, the results of the ongoing TRINOVA-2 and TRINOVA-3 trials of trebananib should become available within the next few years, and these results should provide a better understanding of how trebananib will best fit into the treatment paradigm for ovarian cancer.
By: Stephanie Hawthorne, PhD, Director, Clinical and Scientific Assessment, Kantar Health and Cory Blaiss, Analyst, Clinical and Scientific Assessment, Kantar Health
The melanoma field exploded two years ago, with the 2011 and 2012 approvals of Yervoy® (ipilimumab, Bristol-Myers Squibb) and Zelboraf® (vemurafenib, Genentech/Roche/Daiichi Sankyo). The approval of Zelboraf also ushered in the era of personalized medicine in melanoma, effectively segmenting the market in half – those patients with BRAF V600E mutation (approximately 44% of patients), and those without. In May 2013, the field expanded again following the FDA approvals of Tafinlar® (dabrafenib, GlaxoSmithKline) and Mekinist® (trametinib, GlaxoSmithKline), and one month later the European Medicines Agency (EMA) approval of Tafinlar, for treatment of BRAF-mutant unresectable or metastatic melanoma.
According to Kantar Health’s CancerMPact® Treatment Architecture 2013 data, which was surveyed prior to the recent approvals of Tafinlar and Mekinist, 77% of U.S. and 48% of Western European BRAF-mutant metastatic melanoma patients receive Zelboraf as first-line therapy. However despite impressive responses and improved outcomes, resistance to these agents develops and relapse occurs within six months. In these resistant patients, reactivation of the RAF-MEK-ERK pathway and mutations in MEK have been documented. These observations, coupled with preclinical examination of the hypothesis that targeting both BRAF and MEK simultaneously could provide greater efficacy, more durable responses and reduced toxicity, has led to testing of the combination of BRAF and MEK inhibitors in melanoma patients.
At the 2013 American Society of Clinical Oncology (ASCO) annual meeting, we saw data from a Phase I/II trial of the combination of BRAF and MEK inhibitors (Sosman, Abstract 9005), Tafinlar and Mekinist. The combination had impressive objective response rates (63-76% ORR including 8-9% complete responses, CR) in BRAF inhibitor-naïve patients, nearly additive of the ORRs for the monotherapies (50% for Tafinlar and 23% for Mekinist). The PFS ranged from 9.2 to 9.4 months across all dose cohorts (Flaherty et al., NEJM, 2012). Based on this promising data, we expect the combination data to shift market share away from single agents to the combination of Tafinlar and Mekinist. Indeed, this is already occurring, despite lack of approval for the combination therapy. In late June 2012, Kantar Health surveyed 30 U.S. community oncologists about their preferred choice of first line therapy for BRAF-mutant metastatic melanoma patients. The results show nearly one-fifth of patients receiving the Tafinlar-Mekinist combination off-label.
Genentech/Roche is hoping to capitalize on this trend by testing its own BRAF and MEK inhibitor combination. Updated results for the Phase Ib BRIM7 trial were presented on September 28 at the 2013 European Cancer Congress (ECCO-ESMO-EORTC) annual meeting (McArthur, Abstract 3703), examined the combination of Zelboraf (BRAF inhibitor) and cobimetinib (MEK inhibitor) in 63 Zelboraf-naïve and 65 Zelboraf -resistant patients. In the BRAF inhibitor-naïve patients the ORR was an impressive 85%, including a 10% CR rate. The median PFS had not yet been reached despite 10 months of median follow-up. This updated data from the original 2012 publication reflects a slight strengthening of the data, with ORR remaining similar but now observing complete responses in this larger data set. In patients who had progressed on prior Zelboraf there was only a 15% ORR, identical to what was observed in the GSK combination study. This lack of significant efficacy when using the combination therapy after progression on BRAF inhibitor therapy is a strong argument for using the combination therapies upfront.
Generally, there were more adverse events with the combinations than with the monotherapies (with the notable exception of less squamous cell carcinoma with the combination therapy) and differential toxicity profiles between the two combination regimens. The Tafinlar + Mekinist regimen had higher rates of fever than the Zelboraf + cobimetinib combination (71% and 43%, respectively); the Zelboraf + cobimetinib combination had higher rates of diarrhea (all grades: 81% and 36%, respectively; grade 3/4: 8% and 2%, respectively) and rash (all grades: 89% and 27%, respectively; grade 3/4: 13% and 0%, respectively). The incidence of squamous cell carcinoma was similar between the two regimens (5% grade 3/4).
In terms of efficacy it appears that the Genentech/Roche combination has the advantage; however, the GSK combination has a significant first-mover advantage. Not only are both GSK agents already approved in the U.S., but GSK has already received FDA priority review for accelerated approval of its combination based on the randomized Phase II data, with a Prescription Drug User Fee Act (PDUFA) target date of January 8, 2014. In Europe, the pending approval for Mekinist includes a broad label which includes use in combination with Tafinlar. Further, GSK was the first to initiate two phase III trials in mid-2012 – one in comparison with Tafinlar monotherapy (NCT01584648) and one in comparison with Zelboraf (NCT01597908). Genentech/Roche initiated a trial of their combination against Zelboraf (NCT01689519) six months later, in January 2013, putting it two or more years behind GSK. Will the first-to-market advantage favor the GSK combination? Or will physicians, more familiar with Zelboraf, favor the Genentech/Roche combination? It is too early to predict what the phase III data will show, but it will certainly be a showdown.
By: Stephanie Hawthorne, PhD, Director, Clinical and Scientific Assessment, Kantar Health and Mara Jeffress, PhD, Associate Consultant, Clinical and Scientific Assessment, Kantar Health