Lutathera’s Broad Tumor Indication Aided By Expanded Access Data
Executive Summary
US FDA’s approval of Advanced Accelerator Applications’ radiopharmaceutical for gastroenteropancreatic neuroendocrine tumors was driven by robust Phase III study results in midgut carcinoid tumors and data from an investigator-initiated, expanded access study in the Netherlands that enrolled patients with a host of other neuroendocrine tumor types.
Amid the ongoing US debate about whether patients should have a statutory right to try investigational drugs, expanded access programs and emergency use requests factored into two recent US FDA approvals in diseases with high unmet needs.
The approvals of Advanced Accelerator Applications SA's Lutathera (lutetium Lu 177 dotatate) for gastroenteropancreatic neuroendocrine tumors (GEP-NETS) and La Jolla Pharmaceutical Co.'s distributive shock treatment Giapreza (angiotensin II) show how expanded access programs can help move drugs across the regulatory finish line more quickly, or for a broader patient population.
In the case of Lutathera, FDA concluded data from a long-term, investigator-initiated, expanded access protocol in the Netherlands provided support for a broader indication than otherwise would have been justified by the lone pivotal trial, even though there were significant limitations with the data gathered through the Dutch study.
FDA highlighted the role of expanded access data in a Jan. 26 press release announcing the approval of Lutathera as the first radiopharmaceutical for the treatment of GEP-NETs, a rare group of cancers with limited treatment options after failure of initial therapy.
“This approval provides another treatment choice for patients with these rare cancers. It also demonstrates how the FDA may consider data from therapies that are used in an expanded access program to support approval for a new treatment,” said Richard Pazdur, director of the Oncology Center of Excellence and acting director of the Office of Hematology and Oncology Products.
For Giapreza, emergency use requests following publication of Phase III results in the New England Journal of Medicine spurred FDA to an early approval, cutting two months off the review clock. (See sidebar for story.)
FDA often encourages sponsors to open expanded access protocols for investigational drugs in late-stage development for serious diseases. The Lutathera and Giapreza approval experiences suggest FDA saw not only the utility of the expanded access data, but also the limits that such programs have in making promising drugs broadly available to patients who need them.
Poor Quality Submission Leads To CRL
Lutathera is a radiolabeled somatostatin analog under development by BioSynthema Inc. before the company's acquisition by Advanced Accelerator Applications (AAA) in 2010. In October 2017, Novartis AG announced plans to acquire AAA, and that deal closed just days ahead of FDA’s approval of Lutathera.
The drug had a long development history, with BioSynthema and AAA targeting approval for treatment of somatostatin receptor-positive GEP-NETs.
Neuroendocrine tumors are a rare, clinically diverse group of epithelial malignancies that originate most commonly from the gastrointestinal tract and pancreas, according to FDA’s multidisciplinary review. Tumors are principally categorized by the level of differentiation and may be further characterized as functional or nonfunctional depending on whether they secrete peptide hormones that result in clinical symptoms.
The majority of well-differentiated GEP-NETs express somatostatin receptors (SSTRs) and can be imaged using a radiolabeled form of the somatostatin analog octreotide (Novartis’ Sandostatin LAR Depot), the review states, adding that treatment of functional GEP-NETs requires management of clinical symptoms in addition to anti-cancer therapy.
The original submission resulted in a complete response letter due to a host of database quality problems with the application, including data that were materially incomplete, inaccurate, untraceable, and inconsistent.
Systemic therapy options for well-differentiated, metastatic GEP-NETs include somatostatin analogs such as octreotide and lanreotide (Ipsen's Somatuline Depot), targeted therapies such as sunitinib (Pfizer Inc.’s Sutent), and chemotherapy.
In a March 2011 pre-IND meeting to discuss BioSynthema’s proposed randomized trial, FDA stated that while the subpopulation of patients with midgut carcinoid tumors was acceptable for study, given the biological and clinical heterogeneity of GEP-NETs it would be unlikely that a broad indication for the treatment of somatostatin receptor-positive GEP-NETs would be granted based upon data derived from the proposed trial. The agency recommended the sponsor conduct two adequate and well-controlled trials.
Ultimately, the sponsor conducted only one pivotal trial but sought to rely on data from an expanded access protocol in the Netherlands for additional support.
In a rolling NDA submission completed in April 2016, AAA sought approval for treatment of adult patients with somatostatin receptor-positive GEP-NETs, including those of the foregut, midgut, and hindgut. This original NDA submission resulted in a December 2016 complete response letter due to a host of database quality problems with the application, including data that were materially incomplete, inaccurate, untraceable and inconsistent. (See sidebar for timeline.)
"Complete, accurate, reviewable, transparent, traceable, and robust data are the foundations of any successful NDA,” FDA’s first-cycle multidisciplinary review states. “The amount of confidence reviewers have in these data attributes is the amount of confidence reviewers are able to have in the strength of the application. The information provided in NDA 208700 is of insufficient quality and integrity to be able to confirm the effectiveness of Lutathera for the proposed treatment population."
FDA noted the decision to issue a complete response letter "was reached after considerable efforts were made to resolve the issues with the applicant."
NETTER-1 Convincing In Midgut Tumors …
AAA resubmitted the NDA in July 2017; the application was largely based on data from the NETTER-1 trial and the expanded access protocol from Erasmus Medical Center in the Netherlands.
NETTER-1 was an open-label, randomized study that enrolled 229 patients with advanced/inoperable or metastatic SSTR-positive midgut carcinoid tumors that had progressed while on octreotide LAR. Patients were randomized to Lutathera or octreotide LAR. The primary endpoint was progression-free survival (PFS) based on an independent review assessment; overall response rate (ORR) and overall survival were major secondary endpoints.
Lutathera demonstrated a statistically significant and clinically meaningful treatment effect, with the estimated median PFS not reached for patients in the Lutathera arm compared to 8.5 months in the octreotide LAR arm (HR 0.21 [95% CI: 0.13, 0.32], p<0.0001).
Patients who received Lutathera demonstrated a statistically significant improvement in ORR compared to those in the octreotide arm (12.9% and 3.5%, respectively). The estimated median overall survival for an updated analysis requested by FDA was not reached in the Lutathera arm and was 27.4 months in the comparator arm (stratified HR 0.52 [95% CI: 0.32, 0.84)], p=0.0068). The survival analysis was immature, and a statistically significant improvement was not demonstrated for failure to cross the conservative statistical boundary set for the interim analysis.
Despite certain clinical and statistical limitations with NETTER-1, "approval based on this trial should be granted because of the large treatment effect that overcame these limitations," Steven Lemery, associate director of the division of Oncology Products 2, said in his review. (See sidebar for reviewers.)
While FDA reviewers were persuaded of Lutathera's efficacy in midgut tumors, they were not convinced based on the sole pivotal trial that a broad indication in GEP-NETs was warranted. Consequently, they looked to data from the Erasmus Medical Center (EMC) study.
… But Erasmus Data Lend Support For Other GEP-NETs
Right-to-try legislative proposals that have been floated in Congress during the past year include some that would take FDA out of the equation when it comes to patient requests for compassionate use of investigational drugs and limit its ability to consider safety events occurring during expanded access in its evaluation of a drug's benefit/risk profile. Such legislation is opposed by industry and many patient advocacy groups, but nevertheless appears to be nearing passage, with the House preparing for a decisive vote which could come as soon as Tuesday.
In a bid to counter the right-to-try movement, there appears to be growing momentum for designing formal expanded access protocols to serve as data-gathering tools to learn more about investigational drugs' efficacy and safety. (Also see "Expanded Access Programs Eyed For Data-Gathering Purposes" - Pink Sheet, 18 Apr, 2018.)
The Erasmus experience shows that FDA can see its way to using both efficacy and safety data from expanded access protocols even when they were not set up with data-gathering intent and there are significant limitations to the data.
However, the Erasmus experience shows that FDA can see its way to using efficacy and safety data from expanded access protocols even when they were not set up with data-gathering intent and there are significant limitations to the data.
The EMC trial was an investigator-initiated, open-label, single-arm, single-institution retrospective study of 1,214 patients with SSTR-positive NETs conducted from January 2000 through December 2012. EMC’s Medical Ethical Committee allowed the treatment of patients with Lutathera on a compassionate basis.
Two-thirds of the patients enrolled were from the Netherlands, and a total of 360 patients (30%) had GEP-NETs. "The study population was heterogeneous with respect to primary tumor site," FDA's second-cycle multidisciplinary review states. "Most patients had GEP-NETs of the foregut, midgut, hindgut, digestive tract, bronchus, and pancreatic neuroendocrine tumors (pNET).”
Patients were treated using an institutional protocol derived from the study of other high-dose radiolabeled peptides conducted at EMC using a different commercial product. The major efficacy outcome was investigator-assessed ORR.
In meetings with the sponsor dating back as far as 2011, FDA repeatedly expressed reservations about the limitations of the Erasmus data to support the broader indication for treatment of GEP-NETs. These concerns included:
- No formal clinical protocol or prespecified statistical analysis plan;
- Baseline tumor assessments were obtained for only two-thirds of EMC study patients;
- Patients were not required to have disease progression following the most recent prior therapy for NET;
- There was no central review of histology or of OctreoScan uptake at baseline or of response;
- There was a high loss to follow-up among non-Dutch participants, and a differential reporting of serious adverse events between Dutch and non-Dutch participants;
- Prospective data collection was limited and was supplemented by retrospective medical records review and data verification only in the Dutch subset of patients;
- Follow-up assessments measured by CT or MRI were obtained less regularly and less frequently than in the NETTER-1 trial; and
- Two different response criteria were used during the study: the Modified Southwest Oncology Group (SWOG) criteria were used to assess patients who enrolled prior to March 15, 2007, after which patients were assessed using Response Evaluation Criteria In Solid Tumors (RECIST).
“The above limitations concerning ERASMUS suggest that despite its size, the conclusions drawn from this trial concerning the efficacy and safety of Lutathera must be interpreted with caution,” the multidisciplinary review states.
FDA conducted a sensitivity analysis stratified by date of investigator response assessment using SWOG criteria converted algorithmically to RECIST criteria (patients before March 15, 2007) and investigator assessment using RECIST (patients on or after March 15, 2007).
“The analysis of the two populations of patients demonstrated consistently higher estimates of ORR in the population assessed using SWOG criteria and converted to RECIST algorithmically,” the multidisciplinary review states. “Limiting the analysis to patients evaluated on or after March 15, 2007, provided a more conservative estimate of benefit in this trial.”
There were 58 responders in this latter subgroup, and the estimated response rate was 16%, with a median duration of response of 35 months.
“Although from a subset of patients treated, these data provide statistically conservative estimates that were verifiable and are clinically meaningful for patients with GEP-NET,” the review states. “The results provide additional support for the indicated population that are consistent with the observed benefit in other populations of patients with the disease (i.e., in NETTER-1), the biology of the disease itself, the mechanism of action of lutetium Lu 177 dotatate, and the limited treatment options available for these patients.”
Erasmus Contributed Large Safety Database
The Lutathera safety database consisted of data from three sources: NETTER-1, Erasmus, and AAA’s own expanded access/compassionate use programs.
The NETTER-1 trial included 133 patients who received Lutathera. Although the Erasmus database contributed significantly more patients, FDA noted a number of limitations with these safety data.
Only serious adverse events were reported, and there was no active follow-up of non-Dutch patients conducted at EMC. “Furthermore, only data from the Dutch subset was verified against source documents in the retrospective review. Thus, for estimation of the risk of serious adverse events, the Dutch subset of Erasmus is likely to more accurately reflect the true risk,” the multidisciplinary review states. However, the Erasmus study provided an opportunity to assess risks at dose ranges above AAA’s proposed dose, the agency said.
FDA reviewed the safety data from Erasmus “to identify sentinel, rare adverse events and adverse events of special interest not identified in NETTER-1, acknowledging that estimates derived from this trial may underestimate the true risk.”
“With a median follow-up time of more than four years, the most serious chronic toxicities reported were myelodysplastic syndrome (2%), renal failure (2%), cardiac failure (2%), acute leukemia (1%), myocardial infarction (1%), and neuroendocrine hormonal crisis (1%),” FDA said.
AAA also provided data on “a large and diverse population of patients treated under expanded access and compassionate use programs,” FDA’s review states, noting this group totaled 1,695 patients.
“The follow-up time available on these patients was limited and reporting focused on serious adverse events,” the review states. “Review on this patient population focused on screening for sentinel acute and short-term serious adverse events not identified in the NETTER-1 trial.”
Company-Sponsored Expanded Access Programs
AAA’s US expanded access program (EAP) was opened with FDA’s consent and encouragement.
FDA and AAA discussed possible approaches to developing an EAP in the US during an August 2015 meeting on the top-line results from the NETTER-1 PFS analysis.
“AAA proposed to submit a plan for expanded access in anticipation of the public disclosure of the data,” FDA’s meeting minutes state. “FDA strongly encouraged this approach and suggested that AAA submit a treatment protocol for 177Lu-DOTA0-Tyr3-Octreotate for patients with mid-gut carcinoid tumors (the population studied in the NETTER-1 trial) with limited data collection (primarily serious adverse event reports).”
Through February 2018, 348 patients had been treated with Lutathera in the US expanded access program, while almost 2,000 patients in Europe had received the drug through compassionate use/named patient programs.
“For other NET, FDA recommended that AAA make their drug available to investigators under single patient INDs, pending determination of the interest in such access,” the minutes state.
The Lutathera US EAP was opened in March 2016 for patients with inoperable, progressive, somatostatin receptor-positive midgut carcinoid tumors. In July 2017, FDA allowed an expansion of the US EAP to include patients with inoperable, somatostatin receptor-positive neuroendocrine tumors that progress under somatostatin analogue therapy, AAA told the Pink Sheet.
As of the end of February 2018, a total of 348 patients had been treated with Lutathera through the US EAP, while almost 2,000 patients in Europe had received the drug through compassionate use/named patient programs, the company said.
Following FDA’s Jan. 26 approval of Lutathera, new patients are not being admitted into the US EAP, but any patient that was already receiving treatment under the program prior to approval will receive their full course of therapy, according to the company.