St. Jude Children’s Research Hospital (Tennessee, USA) recently opened a trial for patients newly diagnosed with high-risk neuroblastoma (trial number NCT01857934/NB2012, “Therapy for Children with Advanced Stage Neuroblastoma”). This new trial is being led by pediatric oncologist, Dr. Wayne Furman. The following is an email conversation between NB Globe and Dr. Furman about this new trial.
St.Jude’s new trial uses ANBL0532 as the basis of the treatment and builds upon it by adding multiple cycles of immunotherapy using hu14.18K322A. The following is a basic overview of this protocol:
Screening Phase:
1. Tests and Evaluation
Induction Phase:
2. Cyclophosphamide, topotecan, and hu14.18K322A with IL-2 (SQ) then GM-CSF
3. Cyclophosphamide, topotecan, and hu14.18K322A with IL-2 (SQ) then GM-CSF
4. Stem cell harvest
5. Surgical resection to “remove as much tumor as possible”
6. Cisplatin, etoposide, and hu14.18K322A with IL-2 (SQ) then GM-CSF
7. Vincristine, cyclophosphamide, doxorubicin and hu14.18K322A with IL-2 (SQ) then GM-CSF
8. Cisplatin, etoposide, and hu14.18K322A with IL-2 (SQ) then GM-CSF
9. Vincristine, cyclophosphamide, doxorubicin and hu14.18K322A with IL-2 (SQ) then GM-CSF
10. High-dose busulfan and melphalan with stem cell transplant
Consolidation/Intensification Phase:
11. Day +2 to +5 after stem cell infusion, the patient is given 4 doses of hu14.18K322A with NK cells. The NK cells are harvested by a donor who is a biological parent and are “given prior to early hematopoietic cell recovery.”
12. Radiation
Maintenance/Minimal Residual Disease Treatment Phase:
13. hu14.18K322A with IL-2 (continuous infusion), GM-CSF
14. Accutane
NB Globe: What is hu14.18K322A?
Dr. Furman: This is a unique humanized monoclonal antibody against a disialoganglioside called GD2 that is uniformly expressed on neuroblasts. This particular monoclonal antibody (mAb) has been engineered to remove the complement binding sequences. Complement binding is thought to be the cause of much of the pain experienced by patients who receive an anti-GD2 antibody.
NB Globe: How is hu14.18K322A different from hu14.18?
Dr. Furman: The antibody currently used by COG is ch14.18 which is a chimeric (~25% mouse and 75% human) mAb. Hu14.18 is identical to ch14.18 except the Fab ends involve fully human amino acid sequences for IgG1 heavy and kappa light chains, and the complementarity determining regions correspond to the antigen binding sequences of the murine 14.18 mAb. The resulting hu14.18 mAb is approximately 98% derived from human genes, yet maintains the specificity of the murine ch14.18 mAb. In addition, hu14.18K322A also has a single point mutation (K322A) reducing complement-dependent lysis. hu14.18K322A is also produced in a differed type of cell which results in decreased fucosylation (adding fucose to a molecule) and this has been shown to increase ADCC (antibody dependent cellular cytotoxicity), which is how all anti-GD2 mAbs are thought to kill neuroblasts.
NB Globe: How are you building on some of St. Jude’s existing trials with this one? Specifically NCT01576692 (GD2NK) and NCT00743496 (SJGD2)?
Dr. Furman: The second study you cite, SJGD2, is our Phase I study of hu14.18K322a, which was designed to determine the maximum tolerated dosage (MTD) of this antibody when given daily for 4 consecutive days, repeated every 28 days. This SJGD2 study determined the MTD of hu14.18K322a to be 60 mg/m2/day when given daily for 4 days. For NB2012, we chose a dose of 40 mg/m2 to give with chemotherapy, one dose-level below the single agent MTD.
The first study, GD2NK, was designed to evaluate whether we could give this dose of hu14.18K322A with chemotherapy or with chemotherapy and natural killer cells (NK) harvested from a parent. The GD2NK study convinced us that hu14.18K322A is tolerable when given with chemotherapy. Therefore in NB2012 we are now able to test whether the addition of this antibody to standard induction chemotherapy (identical to that given in recently closed Children’s Oncology Group, COG, study, ANBL0532) would improve the response rate in newly diagnosed children with high-risk neuroblastoma.
NB Globe: Are there any conclusions and results from NCT00135135 (“Therapy for Children with Neuroblastoma”) which have helped to shape this new trial?
Dr. Furman: In order have the best chance of determining if giving an antibody against GD2 during induction chemo would improve the response rate, we chose to adopt the COG high-risk induction as given in ANBL0532. Hopefully we can better compare our results to the response of a large group of patients given the identical induction chemo, without the antibody.
NB Globe: Does the screening phase include testing for the ALK gene mutation? Is any other genetic testing done (i.e., PHOX2B)? If a patient is positive for the ALK gene mutation, will the protocol change in any way for them?
Dr. Furman: We are developing the test for ALK gene mutation but it is not yet available at our institution. Also there are no data yet on how children will tolerate the ALK inhibitor, crizotinib, with chemotherapy. A trial to test that has just opened within the COG DVL consortium (ADVL1212). By the time the results for this study are available we should have the ALK gene test available and then we will strongly consider giving newly diagnosed children with this mutation different treatment. However as far as we know now, < 10% of children will likely respond to this new agent.
NB Globe: In COG’s upcoming trial (NCT01798004, ANBL 12P1), they will be using only 5 cycles of induction chemotherapy, removing the second round of vincristine, cyclophosphamide and doxorubicin. This is a change from ANBL0532. How was the decision made to leave in this 6th cycle of chemotherapy with your trial?
Dr. Furman: As discussed above, we wanted to be able to compare our results directly to ANBL0532. Also there is no compelling data suggesting 5 cycles is “better” than 6. In fact our previous high-risk protocol, NB20051 had 11 courses of induction prior to consolidation and our long-term disease free survival, so far is similar to that published by COG investigators. The optimal number of cycles is still unknown.
NB Globe: It is possible that there will be many side-effects to manage during the induction phase of treatment – side-effects from the chemotherapies and possibly from the hu14.18K322A. How will all of these be effectively managed?
Dr. Furman: That was the purpose of the GD2NK study previously mentioned. We have given children hu14.18K322A with 3 different chemotherapy combinations and these side effects were effectively managed.
NB Globe: Is the pain experienced by patients getting hu14.18K322A different from patients getting hu14.18? How is pain typically managed for hu14.18K322A?
Dr. Furman: “Different” is hard to define. We manage pain from both antibodies very similarly although subjectively our impression is that patients we have treated with ch14.18 have more pain and other side effects that those we have treated with hu14.18K322A. However this is difficult to directly compare as the patients given ch14.18 are on ANBL0032 and are also getting either high-dose IL2 or GM-CSF. Because of the unique nature of hu14.18K322a (detailed above) we will be evaluating whether patients will be able to tolerate more of this antibody during the minimum residual disease (MRD) phase of treatment in NB2012.
NB Globe: If it is correct that the surgery comes after the two cycles of cyclophosphamide and topotecan, is there any concern that putting the surgical resection after only two cycles of chemotherapy is too aggressive?
Dr. Furman: We have found that about 2/3rds of patients are able to have their primary tumor successfully resected after the initial two courses of therapy2 and we have been doing that for quite some time now. The timing of resection is always individualized depending on response and other factors for each patient. We have found that the risk of surgery changes very little after the first two course of chemotherapy. In fact there is some information suggesting that earlier tumor removal may decrease the chance for the subsequent development of chemotherapy-resistant disease.
NB Globe: During the consolidation/intensification phase of treatment, will patients be treated proactively with defibrotide to help mitigate the possibility of sinusoidal obstruction syndrome (aka VOD)?
Dr. Furman: Defibrotide is currently only available in the context of a clinical trial so we only can use it once a diagnosis of VOD is made.
NB Globe: Research into busulfan and melphalan (BuMel) therapy has shown that there are differences in how the body processes busulfan, which can result in differences in its level of efficacy and the severity of side-effects. Will pharmacokinetic studies be done to determine how to maximize the efficacy of BuMel while minimizing its toxicity?
Dr. Furman: Absolutely. Busulfan pharmacokinetics will be done and adjusted accordingly for every patient by our expert team of clinicians in the Department of Bone Marrow Transplantation and Cellular Therapy.
NB Globe: Could you please describe the treatment phase that involves giving hu14.8K322A with NK cells? How are the NK cells collected from the biological parent? Are they ‘treated’ in any way before being given to the patient? How are the NK cells given to the patient? What are some of the side-effects?
Dr. Furman: NK cells are collected from the biologic parent in the same way someone would donate platelets. The NK cells are not treated just purified from the other blood cells collected and then are administered intravenously to the patient in a large syringe over a very brief time, usually a few minutes. We have been doing this on our GD2NK study, and so far have not been able to identify any toxicities we can attribute to NK cell administration. The rationale for this phase of treatment is as follows: Most information suggests that the NK cells are the major effector cells of all of the different anti-GD2 monoclonal antibodies (mAbs). Immediately post consolidation with BuMel most patients have no NK cells. That is why the start of MRD treatment with ch14.18 has usually been delayed until about 80-100 days after consolidation is finished. We reasoned if we could give healthy effector cells (NK cells from a parent) with the hu14.18K322A mAb during the recovery phase of consolidation, we may be able to prevent some relapses. Currently further systemic treatment is delayed until patients have recovered from consolidation, which includes radiation to the tumor bed. In NB2012 we will try to give the hu14.18K322A beginning about two days after the PBSC (peripheral blood stem cell) infusion and then the parental NK cells will be given on about day +4.
NB Globe: How does hu14.18K322A work synergistically with NK cells?
Dr. Furman: Monoclonal antibodies work through a mechanism called antibody-dependent cellular cytotoxicity (ADCC for short). ADCCis a mechanism of cell-mediated immune defense whereby an effector cell of the immune system (Natural killer cells or NK cells for short) actively lyses a target cell, whose membrane-surface antigens have been bound by specific antibodies (such as hu14.18K322a). As mentioned previously, the major cell of the immune system that works with the hu14.18K322a antibody to lyse neuroblasts is thought to be the NK cell.
NB Globe: Are any cytokines given to the patient after the NK cell infusion (i.e., subcutaneous IL-2)?
Dr. Furman: Yes, exactly. GM-CSF is also given daily.
NB Globe: How will NK cell activity be measured? Are there concerns that the host’s body might reject the donor NK cells? If so, how is this managed? Will you look at KIR ligand mismatch for selecting parents as the NK donor?
Dr. Furman: Various immunologic tests will be done to look at NK cell function and how long these cells last in the host. The host does “reject” the donor NK cells, usually within a week, but that is usually all the time the antibody lasts in the body anyway.
Yes, KIR testing will be done on both parents (if available) and the “best” donor will be selected by Dr. Wing Leung and his team. Dr Leung is one of the world’s experts on use of NK cells in the treatment of cancer.
NB Globe: During maintenance, are G-CSF and GM-CSF both given? This was a little unclear in the trial information.
Dr. Furman: GM-CSF works very similarly to G-CSF to ameliorate febrile neutropenia after induction chemotherapy. However recently investigators from MSKCC published data suggesting GM-CSF improves response to another anti-GD2 mAb, 3F83. Therefore for these reasons we will be using GM-CSF (instead of G-CSF) after every course of chemotherapy for “prophylaxis” against febrile neutropenia.
Our BMT team thinks we will get a better stem cell harvest from patients if we use G-CSF. So only after the course of chemotherapy in which we will attempt a PBSC harvest will both G-CSF and GM-CSF be given.
This phase is identical to ANBL0032 MRD with exception of hu14.18K322A antibody is substituted for ch14.18 used by COG. We will start with roughly same dose of antibody as used in ANBL0032 and if tolerated try to “give more” in successive cohorts of patients.
NB Globe: During maintenance, are there alternating cycles of hu14.18K322A with Accutane like what is done with the COG ANBL0032 protocol? How many cycles of hu14.18K322A will be given in the maintenance phase? How many cycles of Accutane will be given in the maintenance phase?
Dr. Furman: We will be treating patients in this phase of treatment identical to patients enrolled on ANBL0032, with the exception of the mAB used. ANBL0032 uses ch14.8 and we will be using hu14.18K322a. Patients on both protocols will be getting 5 courses of antibody combined with six courses of Accutane.
NB Globe: MYCN amplification is mentioned very specifically a number of times in the participant criteria. Will the results from patients with MCYN amplification be examined separately from those who are not MYCN amplified? Are there any hypotheses or assumptions being made as to how well high-risk MYCN amplified patients may respond to parts or all of this protocol?
Dr. Furman: The COG has a specific risk classification system for newly diagnosed children with neuroblastoma and we will be using the identical risk classification. MYCN is one of the criteria used to determine whether a patient has high-risk disease.
NB Globe: Throughout the trial, will any novel testing and classification be done to evaluate and measure disease? For example, in the upcoming COG trial examining the use of busulfan and melphalan for high-dose consolidation therapy (NCT01798004, ANBL 12P1), real-time curie scoring will be used as well as the identification of an mRNA expression signature.
Dr. Furman: MIBGs will be done periodically and patients will be scored according to this the curie scoring system, by one of the designers of this system, Dr. Shulkin. We will also be evaluating minimal residual disease in bone marrow and peripheral blood.
NB Globe: How many patients will be enrolled in the trial?
Dr. Furman: The primary objective is the response of patients after receiving two courses of hu14.18K322A and cyclophosphamide/topotecan. The chemotherapy is given identical to the first two courses of ANBL0532, with the exception of the addition the antibody. A total of 42 patients are projected to be needed to answer this question.
NB Globe: What do you expect will be the percentage of families who will be traveling for this trial (i.e., not locals, long term home away from home)? Will these families be required to stay at St. Jude for the entire time?
Dr. Furman: We can accept newly diagnosed, untreated children for this trial from anywhere in the US and a limited number from abroad, if they are interested. Unfortunately since hu14.18K322a is still an experimental agent, we can only give it here at St. Jude. Depending on distance from home, many patients will be able to go home in between courses of therapy, if they have a physician willing to help us care for them, if anything should come up while at home.
NB Globe: How many neuroblastoma diagnoses are made at St. Jude each year? How many of these are high-risk patients?
Dr. Furman: We typically see somewhere between 12 -18 newly diagnosed children with neuroblastoma a year and about 40-60% of these would be high-risk. However, now that this study is open we can accept newly diagnosed untreated high-risk patients from outside our immediate catchment area, if the family would want to travel to Memphis to enroll on this clinical trial.
NB Globe: The first six cycles of chemotherapy will have an additional layer of complexity based on the addition of hu14.18K322A and the cycle of hu14.18K322A with NK cells after stem-cell transplant. Induction and consolidation/intensification are two very heavy and difficult phases of treatment. Do you plan on having additional supports (i.e., social work) to help families as they navigate through these two phases in particular since it is possible that they will be even more challenging to navigate?
Dr. Furman: Although this is a very complicated protocol to read because we are trying to answer several important clinical questions, for families, it is not really different than what is recommended for all newly diagnosed high-risk patients in that all patients are treated with all these “phases” of treatment. As described above we have treated relapsed patients with chemotherapy and antibody and have not seen any significant additive toxicity. The pain from antibody administration is usually easily managed with either bolus IV or continuous infusion morphine.
St. Jude has always provided extensive support for families. Every new patient has a social worker assigned. Additionally there are chaplains available. The institution provides housing and meal stipends for a caregiver and patient while they are in Memphis and travel support for one caregiver and patient. There are also apartments available for patients who are a long way from home and need to stay in Memphis for extended periods of time.
NB Globe: If the accrual is completed in 5 years, when will the results be available? How long is follow-up with these patients?
Dr. Furman: We typically follow patients for at least 5 years after therapy is complete.
NB Globe would like to thank Dr. Wayne Furman for his expertise and time for this interview.
References:
1 Furman, W.L., McGregor, L.M., McCarville, M.B., Onciu, M., Davidoff, A.M., Kovach, S., Hawkins, D., McPherson, V., Houghton, P.J., et al. (2012). A Single-Arm Pilot Phase II Study of Gefitinib and Irinotecan in Children with Newly Diagnosed High-Risk Neuroblastoma. Investigational New Drugs, Vol 30, Issue 4, pps. 1660-1670. http://www.ncbi.nlm.nih.gov/pubmed/?term=21796439
2 Davidoff, A.M., Corey, B.L., Hoffer, F.A., Santana, V.M., Furman, W.L., and Shochat, S.J. (2005). Radiographic Assessment of Resectability of Locoregional Disease in Children with High-Risk Neuroblastoma During Neoadjuvant Chemotherapy. Pediatric Blood and Cancer, Vol 44, No 2, pps. 158-162. http://www.ncbi.nlm.nih.gov/pubmed/?term=15481078
3 Cheung, N.K., Cheung, I.Y., Kushner, B.H., Ostrovnaya, I., Chamberlain, E., Kramer, K., and Modak, S. (2012). Murine Anti-GD2 Monoclonal Antibody 3F8 Combined with Granulocyte-Macrophage Colony-Stimulating Factor and 13-cis-retinoic Acid in High-Risk Patients with Stage 4 Neuroblastoma in First Remission. Journal of Clinical Oncology, Vol 30, No 26, pps. 3264-3270. http://www.ncbi.nlm.nih.gov/pubmed/?term=22869886
