Amgen Response to ICER’s Draft Scoping Document on CAR-T Therapy for B-Cell Cancers


Amgen appreciates the opportunity to comment on ICER’s draft scoping document on CAR-T Therapy for B-Cell Cancers. ICER proposes to compare CAR-T therapy to therapies recommended by NCCN guidelines, such as clofarabine, tyrosine kinase inhibitor-based chemotherapy, or blinatumomab in patients ages 3-25 years with relapsed/refractory B-cell acute lymphoblastic leukemia (R/R B-cell ALL).

Pediatric cancer is the number one cause of disease death in U.S. children. An orphan disease, ALL accounts for 75% of all pediatric cancer with the greatest number of cases occurring between 2-5 years of age.1,2 Relapsed/refractory patients face a median survival of 3 months with a limited number of treatment options left to save their lives.3  In this disease, the value of innovative treatments is not so much in question, eclipsed by the far more pressing issue that there are not enough treatments to address the heterogeneity of patient populations.  Oncologists do not face an ‘either-or’ decision but increasingly implement combinations that will address unique characteristics of patients and disease aggressiveness such as performance status, prior therapy, chromosomal/molecular abnormalities, and relapse location.4  No oncologist wants to have a discussion with patients and their families that their cancer is not responding to treatment.  No parent or child wants to hear this and make hard decisions from a shrinking set of treatment options.  

1) It is too early to do an assessment on CAR-T in R/R B-cell ALL in children and young adults; ICER should delay this assessment. The lack of therapeutic options and mature outcomes data, combined with extraordinary patient heterogeneity indicate that it is too early to perform a meaningful assessment. Amgen supports ICER’s stated goal of creating a “more effective, efficient, and just health care system”.5 With an estimated $750 billion in healthcare spend wasted on areas such as unnecessary services and administration costs, ICER has the opportunity to provide insights in areas transformative for US healthcare.6 R/R B-cell ALL is not one of these areas. It is an unintended but inevitable consequence that a premature ICER assessment could result in more hurdles for this patient population where payers and PBMs introduce barriers to treatment access.  This is at an extremely critical time in a young patient’s fight for survival.  In this setting, time and an oncologist’s freedom to choose the right drugs is literally a matter of life and death.  

When adequate data and treatments are available in pediatric R/R B-cell ALL to conduct this assessment, then ICER should consider orphan assessment issues and technical challenges to the proposed approach as outlined in the following points:  

2) ICER should assess CAR-T in pediatric R/R B-cell ALL as an “ultra-orphan” drug: use of ICER’s “non-ultra-orphan” value framework (as applied for common diseases) runs counter to provisions in place to protect these patients.  With only 2,870 new cases each year in the US,7 this rare disease has been externally validated by the FDA’s tisagenlecleucel-t/CTL-019 Orphan Drug and Breakthrough Therapy designation.8,9  Classifying R/R B-cell ALL as a common disease runs counter to the Orphan Drug Act, the very law that was put in place to protect these patients.10 Rather than attempting to extrapolate what the future treatment population might be, ICER should inform its assessment decisions based on data that are currently available. 

3) Any trial comparison is fundamentally confounded by severe selection bias, and would lack validity: ICER should remove blinatumomab as a comparator from the CAR-T draft scoping document and compare within the class of CAR-T therapies.  An apples-to-oranges comparison of CTL-019 in the ELIANA study11 vs. blinatumomab in the MT103-205 study12 will highlight conclusions that are essentially non-existent:

  • The unique requirement for patients to wait for manufacturing of the CAR-T after enrolment to the CAR-T studies meant patients were excluded from the reporting of results because they died, were too sick to continue on study or their CAR-T manufacturing process failed. This alone constitutes a major selection bias. 85% of CAR-T patients also received an additional line of active “bridging” chemotherapy.  How patients respond to this treatment prior to CAR-T infusion introduces definitive patient selection bias that prevents any comparison between the study results of the two treatments. 
  • Beside this definitive selection bias, patient characteristics of both single-arm studies are heterogeneous and cannot be adjusted: indirect treatment comparison using single-arm studies is associated with an unknown amount of bias. The studies used different inclusion / exclusion criteria and baseline patient characteristics are different on several key prognostic factors. Techniques to address this are infeasible due to the paucity of studies and small study sample sizes in pediatric R/R B-cell ALL.


1) ICER should wait to assess treatments in pediatric R/R B-cell ALL until Phase III/confirmatory studies and a greater number of treatment options are available.

It is too early to perform a value assessment in R/R patients: not enough data are available on these treatments at a stage in a patient’s cancer where there is a shrinking set of treatment options. 

  • Having failed the initial line of therapy, R/R B-cell ALL patients face a median overall survival of 3 months.13 Patient heterogeneity is extensive and its relationship to response is poorly understood.14 Oncologists understand well that a patient’s disease is killing them but they do not have enough information on why.  On top of this, the therapeutic armamentarium of a handful of drugs is too small to address this huge patient and disease variation.  Moreover small numbers of patients mean in the first 2 years after FDA approval, far too little is known about new treatments (e.g., type of patients the treatment is best for, mechanism of action) to put one-size-fits all value-based guidelines that second-guess a hematologist-oncologist and the very individual n-of-one cancers they treat.15

A premature assessment based on inadequate evidence could result in delayed treatment access

  • As a society, U.S. children with cancer have special protected status and it is unacceptable to impose treatment barriers to childhood patients and their parents.  This is particularly important for ALL, a leading disease-related cause of death for children aged 1-19.16  With few remaining treatment options for R/R B-cell ALL patients, a premature assessment and pricing threshold for a new treatment would not result in lower prices but in delayed access to treatment.  This could cause serious harm to the very patients it aims to help.  It could lead to stress and additional time for parents, robbing them of valuable time with their sick child.17  More importantly, with a child’s survival at risk, there is no time to wait for pre-authorizations or step-therapy from insurers, which could likely result from a premature assessment.

A premature assessment in pediatric B-cell ALL would threaten first-in-child drug development.

  • Few drugs are approved in pediatric cancer due to challenges in drug development.18 In the 20 years before the enactment of the Creating Hope Act,19 the FDA approved only two drugs developed expressly to treat a pediatric cancer:  Erwinase and Clofarabine.  Even with current laws and policies, only 28 formal labeling approvals involved pediatric oncologic indications.20
  • Measures are needed to address the lack of access that seriously ill children have to novel, unapproved drugs.  Children should benefit first from the advances of science such as encouraging “first-in-child” drug development.  A premature assessment in R/R B-cell ALL, based on inadequate or incomplete data, puts this at risk.

2) ICER should assess CAR-T in R/R B-cell ALL as an “ultra-orphan” drug; use of ICER’s “non-ultra-orphan” value framework (also applied for common diseases) runs counter to societal provisions developed to protect these very patients. 

ICER’s proposed “non-ultra-orphan” designation for CAR-T is in direct conflict with legislation designed to encourage pediatric cancer drug development.

The FDA has defined R/R B-cell ALL as an orphan disease and CTL019 received Breakthrough Therapy and priority designation for this indication.21, 22 ICER’s inclusion and exclusion of what is considered “ultra-orphan” conflicts with the Orphan Drug Act, the law developed to protect and ensure these patients have access to therapies which treat life-threatening conditions. 23, 24, 25, 26 R/R B-cell ALL treatments should be considered and assessed as “ultra-orphan” drugs whether or not these therapies treat other diseases.

3) Technical assessment issues: ICER should not conduct an assessment of CAR-T therapy in pediatric patients with R/R B-cell ALL with blinatumomab as a comparator.

The CAR-T reinfusion waiting period, unique to CAR-T, in which patients receive an additional line of active chemotherapy, introduces definitive patient selection bias that prevents any comparison between the two study results. 

  •  A major difference between the two trials is that in ELIANA, the majority of patients used active “bridging chemotherapy” prior to CAR-T infusion. Patients who died, were too sick to continue on study or whose CAR-T manufacturing process failed between enrollment and reinfusion were excluded from the analysis.  This constitutes a considerable selection bias between the two studies. MT103-205 investigating the “off the shelf” immunotherapy technology, blinatumomab, ensured patients were infused upon enrollment, without “bridging therapy”.
  • Moreover, in ELIANA, because of the need for bridging chemotherapy a proportion of patients actually even achieved a treatment response (rendering them MRD +, not R/R) before receiving their CAR-T infusion. This definitive selection bias prevents any comparison between the two study results.

The heterogeneous populations in ELIANA and MT-103-205 make indirect comparisons infeasible.  Important prognostic factors (age, previous treatment history, baseline blast levels)27 differed between the two trials.

  • Age differences: ELIANA enrolled older pediatric patients up to 3 to 21 years of age while the MT103-205 enrolled younger patients from 0 to 18 years old.
  • Baseline blast level differences: ELIANA required baseline blast levels at enrollment to be greater than or equal to 5%, whereas MT103-205 required baseline blast levels to be greater than 25%.  
  • Previous treatment history differences: In ELIANA, the median number of previous lines of therapy was 3.0, meaning that at least 50% of patients had 3 or more prior lines of therapy; whereas in MT103-205, 49 patients (70%) had only 1 or 2 prior therapies. ELIANA contained only 21% patients with refractory disease compared to 56% of patients in MT103-205. ELIANA required patients to have life expectancy longer than 12 weeks, MT103-205 did not have any such requirements.

Given the aforementioned imbalances in patient characteristics, the indirect treatment comparison needs to adjust for all effect modifiers and prognostic factors as noted in NICE’s recent technical support document on methods for population-adjusted indirect comparisons.28 This is extremely challenging given ELIANA and MT103-205 studies are single-armed and of very small sample sizes (68 vs. 70 patients), notwithstanding the need to develop an ‘overlapping’ patient cohort due to different patient characteristics. In addition, the paucity of single-arm trials for blinatumomab makes the inference unstable.  In choosing comparators, ICER should only use CAR-T therapies and not include blinatumomab.


There is a significant unmet need in the ultra-orphan pediatric R/R B-cell ALL population with median overall survival of only 3 months, which is reflected in the granting of Breakthrough Therapy designation for this indication by the FDA. Moreover, it is important to consider the provisions put in place by the Orphan Drug Legislation to protect treatment access for these patients and ensure future innovation.  ICER should consider the above recommendations and delay this assessment until there is adequate data, more treatment options and a deeper understanding of pediatric R/R B-cell ALL in order to inform a meaningful assessment that puts the patient at the center.


  1. Adamson P. et al. Childhood Cancer Research Landscape Report.  Translating Discovery into Cures for Children with Cancer. p.66.  Link
  2. Margolin JE, Rabin KR, Poplack DG.  Chapter 124: Leukemias and Lymphomas of Childhood. in Devita Jr VT, Lawrence T, Rosenberg SA. Cancer: Principles & Practice of Oncology: Annual Advances in Oncology. Lippincott Williams & Wilkins; 2012 Jan 5.  p. 1793.
  3. Jeha S, Gaynon PS, Razzouk BI, Franklin J, Kadota R, Shen V, Luchtman-Jones L, Rytting M, Bomgaars LR, Rheingold S, Ritchey K. Phase II study of clofarabine in pediatric patients with refractory or relapsed acute lymphoblastic leukemia. Journal of Clinical Oncology. 2006 Apr 20;24(12):1917-23.  Table 2. Treatment Outcome. Link
  4. Op. Cit. Margolin et al. pp. 1797-1799.
  5. ICER Website. Accessed August 2017.  Link
  6. McGinnis JM, Stuckhardt L, Saunders R, Smith M, editors. Best care at lower cost: the path to continuously learning health care in America. National Academies Press; 2013 Jun 10.
  7. Op. Cit. Adamson P. et al. 2016p.66.  Link
  8. FDA. Search Orphan Drug Designations and Approvals [Internet]. Food and Drug Administration;[cited 25 Aug 2017. Link
  9. Novartis. CTL019 (tisagenleucleucel): In pediatric and young adult patients with relapsed/refractory B-cell acute lymphoblastic leukemia. U.S. Food & Drug Administration, Oncologic Drugs Advisory Committee July 12, 2017. [Internet]. Food and Drug Administration website;[cited 25 Aug 2017]. Link
  10. U.S. Food and Drug Administration (FDA).  The Orphan Drug Act.  Relevant Excerpts (Public Law 97-414, as amended) Last updated August 2013.  Link
  11. Novartis. CTL019 (tisagenleucleucel): In pediatric and young adult patients with relapsed/refractory B-cell acute lymphoblastic leukemia. U.S. Food & Drug Administration, Oncologic Drugs Advisory Committee July 12, 2017. [Internet]. Food and Drug Administration website;[cited 25 Aug 2017]. Link
  12. Von Stackelberg A, Locatelli F, Zugmaier G, Handgretinger R, Trippett TM, Rizzari C, Bader P, O’Brien MM, Brethon B, Bhojwani D, Schlegel PG. Phase I/Phase II study of blinatumomab in pediatric patients with relapsed/refractory acute lymphoblastic leukemia. Journal of Clinical Oncology. 2016 Oct 3;34(36):4381-9.
  13. Op. Cit. Jeha S. et al., 2006  Link
  14. Patient heterogeneity is too extensive to make any accurate value assessments, echoed by oncologists such as Dr. Richard Gorlick, Division Chief, Pediatric Hematology/Oncology, The Children’s Hospital at Monte ore.,The sample size is too small. We can’t see a pattern with 80 samples because the cancer is too complex, perhaps we could see a pattern if our n was 1,000, but we can’t get there with pediatric cancers,” in Op. Cit. Adamson P. et al.2016p.13.  Link
  15. Moreover, understanding of chromosomal and molecular abnormalities determining prognosis is shifting so rapidly that ICER’s assessment will be incomplete and outdated even in the 8 months to finalization. See NCCN Clinical Practice Guidelines in Oncology. Acute Lymphoblastic Leukemia.  Version 1.2017.  June 2017.  p. MS7
  16. NIH. NIH Website. National Cancer Institute.  Cancer in Children and Adolescents.  Link
  17. Rosenberg AR, Dussel V, Kang T, Geyer JR, Gerhardt CA, Feudtner C, Wolfe J. Psychological distress in parents of children with advanced cancer. JAMA pediatrics. 2013 Jun 1;167(6):537-43.
  18. Challenges include the small size of the population, biological /genetic differences between children and adults and the potential long term health impact due to cancer and therapies.
  19. Congress created two programs to promote pediatric research on drugs that are otherwise developed for adults.  1) The Best Pharmaceuticals for Children Act (BPCA) provides incentives for drug sponsors to conduct research using their drugs for childhood diseases.  The incentive is in the form of an extra six months of market exclusivity for the drug being tested. 2) Congress passed the Creating Hope Act in 2011 which created a priority review voucher program.   The vouchers are awarded to newly formulated drugs that treat any rare disease in children (not just childhood cancers). A priority review voucher entitles a company to obtain a short FDA review time cutting it from 10 months to six months.  The Advancing Hope Act replaced the Creating Hope Act when it expired in 2016.
  20. FDA approvals from 1953 to 2015. 
  21. FDA. Search Orphan Drug Designations and Approvals [Internet]. Food and Drug Administration;[cited 25 Aug 2017. Link
  22. Novartis. CTL019 (tisagenleucleucel): In pediatric and young adult patients with relapsed/refractory B-cell acute lymphoblastic leukemia. U.S. Food & Drug Administration, Oncologic Drugs Advisory Committee July 12, 2017. [Internet]. Food and Drug Administration website;[cited 25 Aug 2017]. Link
  23. U.S. Food and Drug Administration (FDA).  The Orphan Drug Act.  Relevant Excerpts (Public Law 97-414, as amended) Last updated August 2013.  Link
  24. Breakthrough designation – Congress directed the FDA to establish another program to expedite the development and review of new drugs under Section 902 of the July 9, 2012 Food and Drug Administration Safety and Innovation Act.  Congress passed the Orphan Drug Act (ODA) in 1983 in order to promote the development of drugs for indications like B-ALL and clearly define orphan drugs as drugs that treat conditions that affect 200,000 or fewer people in the US.   Priority Review was developed under the 1992 Prescription Drug User Act (PDUFA) to ensure that overall attention and resources be given to treatments which treat serious conditions.
  25. The goal of the ODA is to provide incentives for manufacturers to develop drugs in conditions that impact a small number of patients as defined as 200,000 or fewer people in the US.  Benefits such a reduced “user fees”, tax deductions for clinical trials, grants and market exclusivity are necessary to encourage research but have not been sufficient to encourage significant growth and development in the pediatric oncology space.
  26. The Pediatric Research Equity Act (PREA) was made into law to increase the number of drugs by requiring manufacturers who develop drugs that are made for adults, to test them in children as well.  Under the current statute, oncology drugs classified as orphan drugs are exempt from the PREA requirement. Non-orphan cancers such as lung, breast and prostate afflict adults and not children and PREA only requires pediatric studies of a drug in the same disease, which is being studied. Thus due to the orphan disease and indication-based exemptions, PREA has limited impact on childhood cancer drug development.
  27. Op. Cit. Margolin et al. pp. 1797-1799.
  28. Phillippo DM, Ades AE, Dias S, Palmer S, Abrams KR, Welton NJ. NICE DSU Technical Support Document 18: Methods for population-adjusted indirect comparisons in submissions to NICE. 2016.  Link