Abstract

Introduction
Multimodality treatment is almost always required to control metastatic cancer. The synergistic combination of chemotherapy and radionuclides has the potential to enhance efficacy and minimize toxicity. Chemotherapeutic agents often radio-sensitize tumors to targeted radionuclide treatment, and cytotoxic effects are additive. Biological molecular-targeted agents may also be pro-apoptotic or increase radionuclide-induced tumor cell death by an alteration of DNA repair mechanisms.
Both non-Hodgkin lymphoma (NHL) and neuroendocrine tumors (NET) are typically indolent malignancies that are disseminated at presentation and are then incurable by conventional treatments. However, specific targeting of the CD20 antigen expressed in NHL, or the somatostatin receptors that are over-expressed and up-regulated in NET cells, offers a potential for the control of progressive disease using a multimodality radioimmunotherapy or radiopeptide therapy approach. Although cure remains an unrealistic expectation, the arrest of tumor progression, achievement of durable remission, and enhancement of quality of life, while simultaneously prolonging survival, are now achievable goals. Since relapse is inevitable, new multimodality radionuclide treatments should be repeatable, to achieve renewed, durable response with acceptable toxicity.
Both follicular NHL and metastatic NET are currently incurable, but effective management may achieve survival up to 10 years. In addition to prolonging overall survival (OS), innovative strategies for controlling these cancers should ameliorate symptoms, and must avoid toxicities that degrade the quality of life. They should also be sustainable, practical, affordable, and acceptable to patients and funding agencies. Expensive new cancer treatments may provide improved progression-free survival (PFS) in the metastatic setting, but have no clear OS benefit or improvement in the quality of life. Cancer patients will accept side-effects for even very small advantages in survival, but it is less clear whether they accept such a compromise of the quality of life for delayed recurrence alone.
The indolent nature of follicular NHL and highly differentiated NET renders the clear demonstration of survival advantage challenging when evaluating new treatment approaches, such as multimodality radionuclide therapy. Surrogate markers such as PFS and measurement of objective response rates (ORR) by RECIST criteria on CT/MRI are unreliable for prognosticating OS, 1,2 and novel, predictive metabolic response end points such as 18F-FDG PET/CT in NHL and 68Ga-DOTA-octreotate PET/CT in NET are not yet validated. Multicenter, randomized controlled trials (RCT) of multimodality radionuclide therapy of both NHL and NET are either scarce or nonexistent, and this novel approach for the control of these cancers has been disparagingly referred to by some oncologists as “an evidence-free zone.” Nevertheless, recent reports on early-phase clinical trials of combination chemotherapy and biologicals with tumor-targeted radionuclides are encouraging.
NHL Radioimmunotherapy
The management of follicular NHL was revolutionized by the advent of rituximab anti-CD20 monoclonal antibody immunotherapy in 1998. Although monotherapy with rituximab alone is relatively ineffective with a complete response (CR) of 5% and a duration of response median of 11 months, 3 the combination with standard chemotherapy regimens such as R-CHOP achieves ORR 96% with a prolonged PFS median of 87% @ 18 months. 4 However, the standard 6-cycle R-CHOP regimen is associated with significant toxicity, and many patients are reluctant to undergo repeated chemotherapy on relapse. A novel radionuclide treatment approach to follicular NHL using 131I-rituximab exploits the triad of avid targeting of the CD20 antigen in the malignant B cell, the radiosensitizing effect of rituximab, 5 and the inherent exquisite radiosensitivity of the lymphoma cell. Radioimmunotherapy with 131I-rituximab has been performed over the past 12 years in our institution to achieve an ORR of 67%, a CR of 50%, and a median OS of 32 months in relapsed/refractory NHL without any bleeding or infection. 6 First-line radioimmunotherapy of follicular NHL with 131I-rituximab, in conjunction with 4 cycles of standard rituximab immunotherapy, achieved an ORR of 98% at 3 months with a CR of 76% and a PR of 22%. PFS had not been reached at a median follow-up of 33 months with no clinically evident toxicity. 7
Our multimodality approach to radioimmunotherapy combines radionuclide therapy with 131I-rituximab with full standard dose immunotherapy with cycles of rituximab 375 mg/m2 each week for 4 weeks, followed by maintenance rituximab at 2–3 months for 1–2 years. This regimen should be contrasted with those of commercially available murine anti CD20 monoclonal antibodies, 131I-tositumomab (Bexxar®), and 90Y-ibritumomab-tiuxetan (Zevalin®), which are given essentially as monotherapy, in association with a single low-dose preload of rituximab to improve biodistribution and promote tumor uptake. Induction doses of 375 mg/m2 rituximab given before 131I-rituximab therapy in a study that correlated serum rituximab measured by ELISA with increased effective half-life showed no compromise of clinical efficacy or increase in toxicity of subsequent anti-CD20-targeted RIT. 8 Indeed, it has been hypothesized that the therapeutic response to Zevalin might be further enhanced by a full course of rituximab followed by a single dose of 90Y-ibritumomab tiuxetan. 9 In newly diagnosed follicular NHL, a phase III randomized comparison of standard R-CHOP against 131I-tositumomab RIT-CHOP in 532 patients demonstrated no statistically significant difference in PFS, OS, or serious toxicities 10 ; however, a further study incorporating full maintenance rituximab after 131I-tositumomab RIT-CHOP is now being performed by SWOG. The first report of a chemo-immunotherapy approach followed by both RIT consolidation and rituximab maintenance described fludarabine, mitoxantrone, and dexamethasone (R-FND) with a 90Y-ibritumomab regimen to achieve a CR of 91% and at a median follow-up of 50 months, projected OS was 93% and PFS was 74% in high-risk untreated follicular NHL. 11
Combined 131I-rituximab radioimmunotherapy with R-CHOP chemotherapy has the potential to reduce the number of chemotherapy cycles from 6 to 3, hence ameliorating the toxicity of the CHOP regimen. This form of 131I-rituximab consolidation radioimmunotherapy may also be used to convert chemotherapy-induced PR to CR, emulating a multicenter international Pharma-sponsored randomized, controlled phase III trial of the consolidation of radioimmunotherapy with 90Y-ibritumomab tiuxetan after first remission in advanced follicular NHL. 12 Multimodality radionuclide consolidation therapy combining 90Y-ibritumomab tiuxetan anti-CD20 murine monoclonal antibody with chemotherapy, comprising chlorambucil, CVP, CHOP, or fludarabine combinations, prolonged PFS by 2 years and converted 77% of patients in PR after the standard chemotherapy regimen, to CR, regardless of the type of first-line induction treatment. After a median follow-up of 3.5 years, the addition of 90Y-ibritumomab tiuxetan consolidation radioimmunotherapy to standard chemotherapy had improved PFS from 13.3 to 36.5 months and achieved CR rates of 87% with manageable additional toxicity. 12 Grade III/IV hematological toxicity comprising neutropenia 66.7% and thrombocytopenia 68.8% required growth factor support in 17.6% and platelet infusion in 20.6%. Six patients died of neutropenic sepsis. 12 A standard activity of 14.8 MBq/kg 90Y-ibritumomab tiuxetan was administered to each patient, and the absence of the ability to prescribe an individual radiation-absorbed dose in Gray to hemopoietic marrow, which is routinely performed with 131I-rituximab radioimmunotherapy utilizing quantitative imaging of the 131I gamma emission, may have contributed to the observed toxicity. Nevertheless, this phase III randomized, controlled radioimmunotherapy study proves the efficacy of multimodality radionuclide therapy and shows the way forward in obtaining the level of evidence that is necessary to translate nuclear medicine therapy into routine oncology clinical practice. It also addresses the frustration expressed by Cheson 13 that in NHL therapy we have seen enough trials comparing various combinations of conventional chemotherapy agents in assorted doses, and schedules for lymphomas and resources would be better expended on trials that rationally combine novel agents which affect multiple receptors or pathways and that lead to individualized therapy and the elusive goal of curing follicular lymphoma.
Multimodality multicenter phase III randomized, controlled studies of radionuclide combination with carmustine, etoposide, cytarabine, and melphalan (BEAM) conditioning before autologous hemopoietic stem cell transplantation (ASCT) for aggressive lymphoma have been recently reported. 14,15 Standard dose 90Y-ibritumomab tiuxetan, with a single predose of 250 mg/m2 rituximab, combined with BEAM high-dose chemotherapy, was reported to be safe and possibly more effective than BEAM alone as a conditioning regimen for ASCT for aggressive lymphoma. 14 However, numbers were small, the trial having been stopped after the recruitment of 22 patients due to slow accrual. The results were encouraging, but did not reach formal statistical significance, the 2 year PFS being 59% after Z-BEAM and 37% after BEAM alone. 14 Similarly, no difference in PFS or OS was observed in more than 200 patients who were randomized to receive either standard dose (0.75 Gy to whole body) 131I-tositumomab and BEAM or standard dose (375 mg/m2) rituximab and BEAM conditioning before ASCT for relapsed DLBCL. The 2 year PFS was 47.9% for Bexxar/BEAM versus 48.6% for R-BEAM®. 15 In this large multicenter study, there was no difference in engraftment, and the toxicities were comparable, apart from more mucositis with Bexxar/BEAM.
While the small single-institution study of 131I-rituximab BEAM conditioning before ASCT in salvage therapy of relapsed, refractory NHL reported in this issue 16 was not randomized, the 16 consecutive patients treated over the past 6 years each received two standard doses of 375 mg/m2 predosing of rituximab before 131I-rituximab (0.75 Gy whole body) and standard BEAM, which is in contrast to both the Z-BEAM and Bexxar BEAM regimens. Full dose rituximab immunotherapy does not appear to increase toxicity and has the potential to augment efficacy. 16
Engraftment was achieved in all patients: 15 out of 16 ASCTs achieved a CR, and 1 out of 15 ASCTs achieved a partial response. Twelve out of sixteen patients remained alive and disease free at a median of 44 months (range 4–108 months) post-ASCT.
This small proof-of-concept clinical study was not powered to provide statistically significant outcomes, but there was an impression of improved disease control in transformed follicular and mantle cell lymphoma. 16
NET Radiopeptide Therapy
Lutetium-177-octreotate radiopeptide therapy of NETs commenced at Erasmus Medical Center, Rotterdam, more than 12 years ago, and a seminal study of more than 500 patients reported objective responses with CR 2%, PR 28%, and MR 16%. 17 A survival benefit of 40–72 months from diagnosis was claimed in comparison with historical controls. A preliminary report of the absence of short-term side-effects after the addition of capecitabine in 7 patients undergoing standard 177Lu-octreotate radiopeptide therapy 18 was published in 2008 as a prelude to a multimodality RCT that is currently in progress at the Erasmus Medical Center. Our single-institution physician-sponsored phase II study of 35 patients with inoperable, progressive, differentiated NET also demonstrated that the addition of capecitabine radiosensitizing chemotherapy did not increase the minimal toxicity of standard dose 177Lu-octreotate radiopeptide therapy. 19 Objective responses of 15% CR, 38% PR, and 38% SD were achieved but, more importantly, tumor control and the stabilization of progressive NET were obtained in 92% of those patients. 19 The addition of temozolomide to the combination of capecitabine and 177Lu-octreotate was prompted by reported efficacy of first-line chemotherapy with capecitabine and temozolomide in patients with metastatic pancreatic endocrine carcinomas. 20 This multimodality radionuclide approach in 35 patients followed for a mean of 18 months, reported in this issue, 21 achieved an ORR of 94% in the cohort of pancreatic NETs, which can be favourably compared with the capecitabine temozolomide combination alone, which reported 70% ORR. 20 It should be remarked that the ORR for standard chemotherapy regimens in well-differentiated unresectable metastatic pancreatic NET is 35%–40%. 22
The advent of m-TOR inhibitors, with their biological rationale for clinical application in NET therapy, 23 and promising early clinical data as single agents in prolonging PFS, 24 invited the addition of 177Lu-octreotate radiopeptide therapy in a multimodality approach. The major anticipated advantages are improvement in objective response, demonstrable tumor shrinkage, and minimization of the challenging side-effects of long-term use of mTOR inhibitors. We embarked on a phase I/IIA clinical trial of 177Lu-octreotate at a standard activity of 7.8 GBq for 4 cycles, in a 32 week course, along with everolimus daily for 6 months, in patients with progressive, well-differentiated (Ki-67 <5%) NETs. The toxicity of this combination is manageable but challenging and underlines the importance of close cooperation between oncologist and nuclear physician in an integrated treatment program at a specialized center. This proof-of-concept, single-institution, early-phase clinical trial is in progress, and preliminary results seem encouraging. However, definitive evaluation will require multicenter studies across national boundaries, which will face many regulatory, logistical, and financial hurdles. 25
Regulatory constraints on University Hospital physician-sponsored clinical trials, particularly in the European Union, 26 have imposed long delays in approvals for multimodality radiopeptide NET studies, for example, more than 2 years for 177Lu-octreotate with capecitabine in Milan (Lisa Bodei, pers. comm., 2012) and 5 years for 90Y-octreotate with cetuximab (Martyn Caplin, pers. comm., 2011). The nonavailability of radiopeptide therapy in the United States precludes a coordinated academic investigation of potential combination treatment with FDA-approved chemotherapies, such as capecitabine and temozolomide, and biological agents, such as sunitinib and everolimus, and until this impediment is overcome, the desired high level of evidence from randomized phase III trials of optimal combination radiopeptide therapy of NET will remain elusive. The Consensus Report of the NCI Clinical Trials Planning Meeting of 2011 27 concedes that both 90Y and 177Lu radiopeptides have demonstrated promising activity in patients with NET, but indicts many reported studies for suboptimal methodology, lack of intent-to-treat analyses, and nonstandard end-point definitions. 27 While the NCI Report admits that randomized phase III studies comparing peptide receptor radiotherapy with standard systemic therapy are warranted, it is unlikely that their conception of such trial designs will include cutting-edge multimodality combination radionuclide therapy. The apparent reluctance to incorporate radiopeptide therapy with standard algorithms of NET management is also reflected in ENETS guidelines 2011, 28 which consigns 177Lu-octreotate to consideration only as second-line therapy of NETS, or salvage treatment, when all other therapies have failed. At the recent ENETS Copenhagen meeting 2012, multimodality radiopeptide therapy was not cited as a first-line option. Only 10% of the attendees considered that 177Lu-octreotate or 90Y-octreotate might play a role in the treatment of progressive G1–2 pancreatic NETS with hepatic metastasis, although 16% acknowledged that they may consider radiopeptide therapy as a second-line option. Preliminary data reported in this issue for the subset of patients with progressive metastatic well-differentiated pancreatic NET treated first line with 177Lu-octreotate, capecitabine, temozolomide multimodality radiopeptide therapy in a single-center phase II study showed that objective response was a CR of 18%, a PR of 64% on RECIST 1.1 criteria. 21 This measured objective response is in stark contrast with the absence of a statistically significant objective response in patients with pancreatic NETS who are treated with recently approved agents such as somatostatin LAR, Everolimus, and Sunitinib, 24,29,30 which only show the delayed progression of the disease without tumor shrinkage.
Multimodality Radionuclide Therapy
Most radionuclide treatments are currently given as monotherapy, whereas the conventional approach in medical oncology is a combination of chemotherapy agents, often incorporating immunotherapeutics, biological pro-apoptotic agents, radiation sensitizers, and external beam radiotherapy. The growing array of radiopharmaceuticals should now prompt an exploration of synergistic combination regimens that enhance the efficacy of tumor-targeted radionuclide treatments and minimize toxicity. Such regimens of combination radionuclides, monoclonal antibodies, chemotherapeutic radiosensitizers, and biological molecular signaling modifiers should then be rigorously studied in well-designed, adequately powered clinical trials.
Systemically administered chemotherapeutic agents and radiopharmaceuticals have distinct pharmacokinetic properties that affect their bioavailability, therapeutic activity, and toxicities. Tissue radiation exposure is also affected by the relative decay times of the given radionuclides, with dosage being delivered over prolonged periods. External beam radiation therapy, on the other hand, penetrates tissues and cellular boundaries without a requirement for distribution to blood, tissue, interstitial space, and cellular targets, and radiation exposure times are brief, often only a few minutes. Clinical studies of chemotherapy combined with external beam radiation are, thus, difficult to extrapolate to radionuclide therapy in multimodality regimens.
Combined chemotherapy and radionuclide treatment programs are in their infancy, and both nuclear physicians and oncologists are only just beginning to wrestle with the complexities of integrated treatment options. The potential for overlapping and augmented toxicities should be addressed, and individualized radiation dosimetry should incorporate an estimation of Biological Equivalent Dose (BED), which may be affected by concomitant chemotherapy. Close cooperation between oncologists and nuclear physicians is required to ensure that treatment protocols can be performed on a repeated basis, with procedures to ensure patient safety. The exigencies of chemotherapy cycles are best accommodated by the provision of radionuclide therapy on an outpatient basis as routinely performed for both 131I-rituximab radioimmunotherapy and 177Lu-octreotate radiopeptide therapy at our institution. 31 The practical advantage of outpatient radionuclide therapy is the optimal scheduling of treatments without a requirement for a hospital radiation isolation suite availability, which avoids delays and that is accompanied by a considerable cost saving and enhancement of quality of life. 32,33
Combined modality radionuclide therapy programs require a well-defined protocol with stipulated treatment time lines, including therapeutic start dates and optimal dosage intervals. Strict patient follow-up should be performed for treatment and toxicity assessment and clear-cut dose adjustment procedures must be in place. Regular interim analyses of outcomes should be scheduled to ensure absence of excessive toxicity or treatment failure. Quality of life should be monitored by validated and appropriate instruments. Both the nuclear physician and oncologist should work together to achieve clinically meaningful beneficial effects in the patient as manifested by relief of disease-related symptoms, ability to carry out normal activities of daily living, and prolongation of OS without toxic side-effects. It is not sufficient to establish a statistically significant effect on a biomarker, such as PFS, to have reliable evidence of a clinically meaningful benefit. The challenge today is to develop new approaches for translating scientific discovery into cost-effective and meaningful improvements in cancer outcomes that are robust in both magnitude and application to the real world. 2,34,35
While proof of concept has been demonstrated in small, single-institution, physician-sponsored studies of multimodality 131I-rituximab radioimmunotherapy of NHL and 177Lu-octreotate radiopeptide of NET, the evidence base required by oncologists will require multicenter RCT that are beyond the capability of physician-sponsored studies. It should be remarked, however, that RCT phase II trials have a high risk of being misleading or unreliable, unless the tumor characteristics required for targeted therapy are defined for each patient. 36 Such targets may be characterized by a genetic signature, but a serious flaw in the imagined personalized therapy future of oncology is the underestimation of tumor heterogeneity, not just differences between tumors, but also heterogeneity within an individualized tumor. 37 This heterogeneity is manifest in NET by inconsistent Ki-67 scores in different biopsies and by varying avidity of metastases for 68Ga-octreotate and 18F-FDG on PET/CT tumor uptake in the same patient, which reflects differing degrees of differentiation and, hence, expression of the somatostatin receptor target for radiopeptide therapy. All eligible patients with well-differentiated, progressive, unresectable NETS exhibiting Ki-67 <5% and tumor avidity for 68Ga-octreotate on PET/CT referred for consideration of radiopeptide therapy in the physician-sponsored study reported in this issue 21 were offered, and accepted, 177Lu-octreotate radiopeptide therapy on protocol, in combination with radiosensitizing chemotherapy with capecitabine and/or temozolomide, or biological therapy with everolimus. This 100% acceptance rate should be contrasted with the current adult cancer patient participation rate of less than 5% in clinical trials of novel cancer treatments generally. 38 It is, thus, possible to provide routine 131I-rituximab radioimmunotherapy of NHL and multimodality 177Lu-octreotate radiopeptide therapy of NET, on an elective weekly outpatient basis, on physician-sponsored study protocols.
The encouraging patient outcomes in small, single-center studies await confirmation, contingent on the uptake by Pharma, with funding necessary for a large-scale RCT evaluation of multimodality radionuclide therapeutic approaches for durable control of indolent NHL and NET. This level of evidence appears to be the prerequisite for translation into mainstream oncology practice.
Footnotes
Acknowledgments
The author wishes to thank Jenny Lavin for manuscript preparation and Anna Chiam for data management. No grant funding or Pharma industry support was solicited or received.
Disclosure Statement
No financial or other conflict of interest is declared.
