⁹⁰Y-Ibritumomab Therapy in Refractory Non-Hodgkin's Lymphoma: Observations from ¹¹¹In-Ibritumomab Pretreatment Imaging

MATERIALS AND METHODS

This was a retrospective study (January 2000–July 2006) of 28 patients with NHL who were treated with ⁹⁰Y-ibritumomab for refractory or relapsed disease. The local Institutional Review Board approved this study. The group included 21 men and 7 women, with an age range of 36–85 y (average, 56.9 y). All patients were previously treated with a combination of rituximab (Rituxan; Biogen Idec Inc. and Genentech USA, Inc.) or R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisolone), external radiation therapy, or bone marrow transplantation and had stage III or IV disease.

A diagnostic dose (185 MBq [5 mCi]) of ¹¹¹In-ibritumomab was administered on day 0 of the ⁹⁰Y-ibritumomab therapy protocol. This dose was followed by pretherapy imaging consisting of whole-body planar anterior and posterior images acquired immediately and at 24, 48, and 72 h using an Infinia Hawkeye scanner (GE Healthcare) and interpreted using a dedicated Xeleris workstation provided by the manufacturer (GE Healthcare). The administered therapeutic doses of ⁹⁰Y-ibritumomab (given on day 7 after the imaging dose) ranged from 629 to 1,258 MBq (17–34 mCi) (average, 1,054.5 ± 164.65 MBq [28.5 ± 4.45 mCi]). The patients were preloaded with unlabeled rituximab (250 mg/m²) on days 0 and 7. The protocol used for therapy is presented in Table 1.

The clinical outcomes of the ⁹⁰Y-ibritumomab therapy were compared with the findings of the ¹¹¹In-ibritumomab scans. Response to radioimmunotherapy was determined on the basis of all available clinical and radiographic follow-up, which included clinical notes, anatomic studies, and ¹⁸F FDG PET/CT scans. The response to therapy at 3 mo after ⁹⁰Y-ibritumomab administration was assessed according to the EORTC criteria (3) for PET in 17 patients (60.7%). Determination of response was based on the referring physicians' overall impression from clinical notes. ¹¹¹In-ibritumomab scans were recorded as positive if any known site of disease was identified and negative if no abnormal uptake was noted.

RESULTS

⁹⁰Y-ibritumomab induced objective responses in 22 of the 28 patients (78.6%). A complete response was noted in 9 patients (32.1%), a partial response in 9 patients (32.1%), and a mixed response in 4 patients (14.3%). There were 3 patients (10.7%) with stable disease and 3 patients (10.7%) with disease progression. ¹¹¹In-ibritumomab scans were available for review in all 28 patients: 19 had positive ¹¹¹In-ibritumomab findings and 9 had negative ¹¹¹In-ibritumomab findings. A complete response was noted in 2 of 19 patients with positive findings and 7 of 9 patients with negative findings. A partial response was seen in 7 of 19 patients with positive findings and 1 of 9 patients with negative findings. Disease progression was observed in 3 of 19 patients with positive findings and 0 of 9 with negative findings. The remaining patients had a mixed response or no changes. These findings are summarized in Table 2.

In Figure 1, we present a 60-y-old woman with NHL and a complete response after ⁹⁰Y-ibritumomab treatment. No lesions were seen on the pretherapy ¹¹¹In-ibritumomab scan (images from the 48-h scan are shown). Maximum-intensity-projection images of ¹⁸F-FDG PET (4 mo apart) showed resolution of the lesions noted before therapy. Figure 2 shows a 45-y-old man with NHL and progressive disease after ⁹⁰Y-ibritumomab treatment. Lesions were seen in the abdominal region on the pretherapy ¹¹¹In-ibritumomab scan (images from the 48-h scan are shown). Maximum-intensity-projection images of ¹⁸F-FDG PET scans (4 mo apart) show progression of the lesions noted before therapy.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1. 

A 60-y-old woman with NHL and complete response after ⁹⁰Y-ibritumomab treatment. (A) No lesions are seen on anterior whole-body planar pretherapy ¹¹¹In-ibritumomab scan (images from 48-h scan are shown). (B) Pretherapy (1 mo) MIP image of ¹⁸F FDG PET shows abdominal lesions (arrowheads). (C) MIP image of ¹⁸F FDG PET after therapy (3 mo) is negative for active disease.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2. 

A 45-y-old man with NHL and progressive disease after ⁹⁰Y-ibritumomab treatment. (A) Lesions are seen in abdominal region (arrowheads) on anterior whole-body planar pretherapy ¹¹¹In-ibritumomab scan (images from 48-h scan are shown). (B and C) Pretherapy (1 mo) MIP image of ¹⁸F FDG PET (B) and posttherapy (4 mo) MIP of ¹⁸F FDG PET (4 mo apart) (C) show progression of lesions noted before therapy (arrowheads).

DISCUSSION

The results of our retrospective review suggest that there may be an inverse relation between the amount of disease visible on the ¹¹¹In-ibritumomab pretherapy scans and the response to ⁹⁰Y-ibritumomab treatment, with a higher rate of complete response seen after ⁹⁰Y-ibritumomab in patients with negative ¹¹¹In-ibritumomab findings, and a higher rate of disease progression noted despite therapy in patients with positive ¹¹¹In-ibritumomab findings. This observation contradicts prior reports that argue against giving the therapy for a tumor that has not been shown to take up the agent (4).

A limitation of our retrospective review is that ¹¹¹In-ibritumomab is intended to assess biodistribution of the radiolabeled antibodies, not to be a sensitive method of disease detection. Thus, these observations need to be confirmed in a trial including proper imaging of the target CD20 receptors. One such attempt is a report by Perk et al., who labeled ibritumomab with ⁸⁹Zr and reported the first use in a human subject. Their PET images obtained after injection of ⁸⁹Zr-ibritumomab showed targeting of all known tumor lesions (5).

If we consider radioimmunotherapy as a step beyond immunotherapy of cancer, this step was prompted by the (relative) failure of the latter. The conventional way to explain the failure is a lack of intrinsic killing effect and a lack of penetration into poorly vascularized tumor masses. The addition of a radioactive label (usually a β-emitter) to the antibody would improve both. Radiation has the potential to cure, and the type of radiation used (β-rays) has a sufficient range to overcome the lack of penetration.

All available data, for all forms of disease, suggest that tumor doses are not high enough to have a curative effect and that measured tumor doses do not predict response (6). Yet there is, at least in lymphomas, a curative effect. This discrepancy has preoccupied researchers (7). Some of the magnified effect of what is effectively a relatively small radiation dose can be assigned to the continuous nature of the irradiation by radiolabeled antibodies (8,9). More important, whereas in external-beam irradiation an average dose to the tumor volume as a whole makes sense, the microscopic dose is heterogeneous (10). This heterogeneity may in fact increase the dose at a microscopic level to vital tumor elements. In addition, recent reports indicate a synergistic effect of rituximab followed by radioimmunotherapy, stressing the important role of unlabeled antibodies in the therapy of NHL (11).

Low-grade lymphomas are refractory to most treatments, and each subsequent treatment is less effective (12). Radioimmunotherapy seemed a possibility to improve the treatment options. At present, the most successful (and Food and Drug Administration–approved) radioimmunotherapy agents for lymphomas are anti-CD20 monoclonal antibodies. Rituximab is a chimeric antibody, used as a nonradioactive antibody and to preload the patient when ⁹⁰Y-ibritumomab is used. ⁹⁰Y-ibritumomab is the ⁹⁰Y- or ¹¹¹In-labeled form of ibritumomab tiuxetan. Bexxar (GlaxoSmithKline) is the ¹³¹I-labeled form of tositumomab. Ibritumomab tiuxetan and tositumomab are mouse anti-CD20 monoclonal antibodies. In neither case is the dose to the tumor calculated, because current in vivo imaging methods allow the calculation of only a macroscopic and average tumor dose, probably irrelevant in view of the microscopic heterogeneity. The administered quantity is limited by considerations of toxicity.

Horning et al. reported on a series of 40 patients treated with ¹³¹I-tositumomab (13). The median number of prior treatments was 4. Most patients had low-grade lymphomas (70%). Some had transformed low-grade (25%), and a few had intermediate-grade. Thirty-five (88%) had disease refractory to rituximab. The overall response rate was 68%, with a median response duration of 16 mo (1+ to 38+ mo). A complete response was seen in 33% of the patients. The median duration was not reached. Similar results were reported by Kaminski et al. about a series of 60 patients (14). Again, the patients included in this group received a median of 4 prior treatments. The low, transformed low, and intermediate grades were distributed as 56%, 38%, and 2%, respectively. Overall response and complete response were 47% and 20%, respectively, with median durations of 12 and 47 mo, respectively. When the method was used as the first treatment, the complete response rate was higher than 72% and the 5-y survival higher than 55% (15). In the context of other treatments and outcomes for low-grade B-cell lymphomas, these results are remarkable. Toxicity is primarily hematologic but is rare and reversible with the actual clinical protocols. Interestingly, only a single paper reported on a prospective and randomized study of the additional effect of the radioactive label (16). In that study, with the same amount of antibody in identical regimens but with one arm adding radioactivity, the complete response rate was 33% in the radioactive group and 8% in the unlabeled group. The overall response rates were 55% and 19%, respectively. Improvement will come by making the delivery more efficient by labeling the antibody, already joined to the target (17).

Other groups evaluated ⁹⁰Y-ibritumomab: Conti et al. reported a low rate (0.6%) of altered biodistribution before therapy but did not comment on any role of the diagnostic scan in predicting the response to therapy (18). Visualization of disease on pretherapy imaging was not analyzed by Jacene et al. in a comparison of ¹³¹I-tositumomab versus ⁹⁰Y-ibritumomab in clinical practice (19). However, Gokhale et al. suggested that bulky sites of disease on pretherapy scans could be used to establish a statistical hierarchy of areas likely to show tumor progression and applied this pattern to direct the use of additional external-beam radiotherapy to augment treatment (20). This is similar to the findings of our study that suggest that patients with bulky disease (identified even on pretherapy ¹¹¹In-ibritumomab scans) may require more aggressive management.

CONCLUSION

A higher rate of complete response after ⁹⁰Y-ibritumomab treatment is seen in patients with negative ¹¹¹In-ibritumomab pretherapy findings, whereas a higher rate of disease progression despite therapy is noted in patients with positive ¹¹¹In-ibritumomab findings. This observation, suggesting that patients with bulky disease (identified even on pretherapy ¹¹¹In-ibritumomab scans) may require more aggressive management, needs to be confirmed in larger prospective trials using proper imaging of the target CD20 receptors.