CD70-targeted platform for diagnosing and treating multiple myeloma.

WHAT IS ALREADY KNOWN ON THIS TOPIC

CD70 is an emerging target in multiple myeloma (MM). Various CD70-targeted therapies and imaging agents are in development, establishing the therapeutic potential of these approaches.

WHAT THIS STUDY ADDS

CD70-targeted theranostic platform demonstrates high tumor uptake for imaging and effective radiotherapy in an MM model, significantly improving survival with manageable toxicity. A clinical case confirms the superior lesion-detection capability of [ 18 F]RCCB6 in a patient with MM.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

This platform could provide a new personalized strategy for CD70-positive MM. It justifies clinical translation and future research into optimized regimens, like repeated dosing or alpha-emitters, to improve durability and treat relapsed/refractory disease.

Background

Multiple myeloma (MM) is a hematologic malignancy characterized by the uncontrolled proliferation of plasma cells in the bone marrow, which can lead to destructive bone lesions, renal impairment, anemia, and hypercalcemia. 1 2 Early diagnosis and effective treatment are essential for improving outcomes in patients with MM. Over the years, significant advancements have been made in the treatment of MM, markedly enhancing patient prognosis and quality of life. 3 , 5 However, despite these advancements, challenges such as treatment-related adverse events, drug resistance, and relapse continue to pose significant obstacles, highlighting the necessity for future research aimed at developing more effective and safer therapies. 6

CD70, also known as TNFRSF7, is a member of the tumor necrosis factor superfamily and a type II transmembrane protein. In hematological malignancies, it is overexpressed in B-cell precursor acute lymphoblastic leukemia, diffuse large B-cell lymphoma, and T-cell lymphoma. The overexpression of CD70 can cause immune exhaustion and promote immune evasion, making it closely linked to tumor occurrence, progression, and prognosis. 7 8 Notwithstanding the achievements of B-cell maturation antigen (BCMA)-targeting chimeric antigen receptor T-cells (CAR-Ts), patients with high-risk MM frequently experience relapse. Recent research has identified CD70 as a promising target, as it is notably upregulated in high-risk MM subtypes. Through structure-guided design, a CD27-based anti-CD70 CAR-T has shown superior in vivo expansion and persistence when compared with single-chain variable fragment-based CARs. Epigenetic analysis has uncovered the key transcription factors that drive CD70 expression. To prevent antigen escape, dual-targeting CAR-Ts against CD70 and BCMA have been developed. These findings indicate that CD70-targeting CAR-Ts may provide a powerful therapeutic alternative for patients with high-risk MM who relapse after BCMA CAR-T therapy. 9 These studies suggest that CD70 holds excellent promise as a potential biomarker for MM diagnosis and therapy.

Positron emission tomography (PET)-CT effectively evaluates the distribution and metabolic activity of lesions in both intraosseous and extraosseous manifestations of multiple myeloma. 10 11 Concurrently, targeted radioimmunotherapy (RIT) demonstrates promising potential in the treatment of myeloma due to its unique radiotherapeutic characteristics. 12 , 14 The development of molecular imaging tracers and therapeutic radiopharmaceuticals targeting the emerging biomarker CD70 is of significant clinical importance and may further expand the therapeutic options available. 12 15 16 Among various molecular imaging modalities, immuno-PET imaging combines the specificity of antibodies and antibody fragments with the sensitivity of PET, 17 emerging as a transformative tool for biomarker visualization, cancer diagnosis, patient selection, and response monitoring. 17 Radiolabeled single-domain antibody (sdAb) tracers targeting CD70, like [ 68 Ga]Ga-NOTA-RCCB6 and [ 18 F]RCCB6, enable real-time, high-resolution imaging of CD70-positive tumors. Preclinical and clinical studies have demonstrated that [ 68 Ga]Ga-NOTA-RCCB6 and [ 18 F]RCCB6 outperform traditional tracers, such as fluorine-18 fluorodeoxyglucose ( 18 F-FDG), in diagnostic specificity and sensitivity, enabling precise lesion localization and longitudinal monitoring of therapeutic responses. 1618 , 20 Due to the short circulation time of sdAb that limits the therapeutic potential of sdAb radiopharmaceuticals, we further engineered an sdAb derivative (ABDB6) by fusing RCCB6 with the albumin binder ABD035. 21 In a recent preclinical study, we reported the excellent diagnostic value of [ 64 Cu]Cu-NOTA-ABDB6 across various lymph models, with improved pharmacokinetics and pharmacodynamics, including enhanced tumor uptake and reduced kidney accumulation. 15

For RIT, the radionuclide 177 Lu has been widely used to create effective radiopharmaceuticals with promising potential. 177 Lu is a widely used β-emitting therapeutic isotope, characterized by relatively low emission energy (maximum Eβ of 0.49 MeV) and a limited emission range (maximum 1.6 mm). It delivers sufficient radiation to induce DNA double-strand breaks while minimizing damage to adjacent healthy tissues. 22 23 Leveraging these properties, targeted radioligand therapies using lutetium-177 have achieved notable clinical success. 22 23 Immuno-PET/CT imaging with [ 89 Zr]Zr-DFO-ABDB6, which targets CD70, may further support the potential clinical application of [ 177 Lu]Lu-DOTA-ABDB6. The exploration of the simultaneous theranostic value of CD70 holds translational significance for radiosensitive tumors such as hematological malignancies. In this study, we develop an ABDB6-derived theranostic pair, namely [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6, and assess their diagnostic and therapeutic potential in a preclinical MM model.

Results

Expression of CD70 in human MM samples and MM tumors

We performed CD70 antibody staining on tumor specimens from 18 patients with MM. Of these, 12 patients (66.7%) showed positive CD70 expression ( figure 1A and online supplemental figure 1A ). In contrast, six patients (33.3%) showed negative expression ( figure 1A and online supplemental figure 1B ). Additionally, we analyzed CD70 expression levels across various MM models, revealing that the RPMI-8226 tumor tissue had the highest CD70 expression ( figure 1B ). Therefore, the RPMI-8226 cell line is suitable for CD70-targeted PET imaging and RIT. Moreover, a 61-year-old male patient with MM was enrolled in this study and underwent both [ 18 F]RCCB6 PET/CT and standard 18 F-FDG PET/CT imaging. Results showed that [ 18 F]RCCB6 had high uptake mainly in the kidneys and bladder, with low uptake in the brain, heart, liver, spleen, and intestines ( figure 2A ). Notably, the low background signal of [ 18 F]RCCB6 enables much more precise visualization of lesions in the sternum, ribs, spine, and pelvic bones compared with those seen with 18 F-FDG ( figure 2B ). [ 18 F]RCCB6 detected 42 lesions in this patient with MM, whereas 18 F-FDG detected only 24 lesions ( figure 2C ). The maximum standardized uptake value (SUVmax) of [ 18 F]RCCB6 in MM lesions is significantly lower than that of 18 F-FDG, with values of 5.59±1.14 versus 4.02±2.05 (p=0.01) ( figure 2D ). However, the tumor/bone ratio (TBR) and tumor/muscle ratio (TMR) for [ 18 F]RCCB6 are both significantly higher than those for 18 F-FDG, with TBR values of 3.80±1.93 versus 2.97±0.61 (p=0.002) and TMR values of 12.19±6.20 versus 8.10±1.65 (p<0.001) ( figure 2E ). The positive imaging results of [ 18 F]RCCB6 Immuno-PET/CT in patients with MM can serve as evidence that high CD70 expression exists in some patients with MM and that [ 18 F]RCCB6 Immuno-PET/CT can be used to screen patients suitable for CD70-targeted therapy.

Immunohistochemical staining of the CD70 antigen. ( A ) Representative images of patients with multiple myeloma, categorized as CD70-positive and CD70-negative. ( B ) Representative images of the CD70 antigen in RPMI-8226, MM1.s, and U266 tumors.

[ 18 F]RCCB6 and 18 F-FDG PET/CT images of a patient with MM. ( A ) MIP and fused PET/CT images of [ 18 F]RCCB6, and ( B ) MIP and fused PET/CT images of 18 F-FDG. Lesions in the sternum, left anterior ribs, and right acetabulum (indicated by arrowheads) were visualized by [ 18 F]RCCB6, while they were not detected by 18 F-FDG. A comparative analysis of [ 18 F]RCCB6 and 18 F-FDG PET/CT imaging in the patients with MM is presented, showing the number of lesions ( C ), SUVmax ( D ), TBR ( E ), and TMR ( E ) for all lesions. Data are expressed as mean±SD. **p<0.01, ***p<0.001. 18 F-FDG, fluorine-18 fluorodeoxyglucose; MIP, maximum-intensity projection; MM, multiple myeloma; PET, positron emission tomography; SUVmax, maximum standardized uptake value;TBR, tumor/bone ratio; TMR, tumor/muscle ratio.

Development and characterization of [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6

ABDB6 is a nanobody derivative capable of binding to human CD70 and human/murine albumin. 15 We developed two ABDB6-derived radiopharmaceuticals, [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6, with radiochemical yields of 75.6% and 92.5%, respectively. Moreover, the radiochemical purities of these compounds exceeded 95% ( online supplemental figure 2A and figure 3A ). The radiochemical purities of [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6 also stayed above 95% after incubation in phosphate-buffered saline (PBS) for 48 hours ( online supplemental figure 2B and figure 3B ).

iTLC analysis and in vitro analysis of [ 177 Lu]Lu-DOTA-ABDB6. ( A ) iTLC analysis of [ 177 Lu]Lu-DOTA-ABDB6, demonstrating greater than 95% purity. ( B ) In vitro stability of [ 177 Lu]Lu-DOTA-ABDB6 after incubation in PBS at 37°C for 4, 12, 24, and 48 hours, respectively. PBS, phosphate-buffered saline.

[ 89 Zr]Zr-DFO-ABDB6 immuno-PET imaging and biodistribution studies in RPMI-8226 xenografts

[ 89 Zr]Zr-DFO-ABDB6 immuno-PET imaging results in RPMI-8226 tumor-bearing mice at various time points from 4 to 120 hours are shown in figure 4A . The data reveal that the RPMI-8226 tumor had higher SUVmax values at 4 hours (1.87±0.58), followed by a steady increase at 12 hours (1.99±0.49) and 24 hours (2.14±0.49), reaching a peak of 2.44±0.52 at 72 hours, before declining at 120 hours ( figure 4B and online supplemental table 1 ). Liver uptake peaked at 4 hours post-injection (2.99±0.63) and decreased by about 40% at 12 hours (1.93±0.50). Conversely, kidney uptake reached its maximum at 24 hours (4.86±1.07) post-injection and decreased by approximately 20% at 120 hours (3.81±0.78). The TMRs were 5.13±2.28 at 4 hours, steadily increasing to a peak of 10.63±5.71 at 72 hours, then declining to 8.39±0.98 at 120 hours ( figure 4C ). In vivo tumor targeting of [ 89 Zr]Zr-DFO-ABDB6 was quantified through biodistribution studies in RPMI-8226 tumor-bearing mice at 4, 24, 72, and 120 hours post-injection ( figure 4D and online supplemental table 2 ). Consistent with the data derived from microPET, significant tumor uptake was observed at 4 hours (4.23±0.37 %ID/g), increasing steadily until 24 hours (7.19±1.31 %ID/g), peaking at 72 hours (7.81±0.79 %ID/g), and then decreasing thereafter. Liver uptake peaked at 4 hours (5.67±0.78 %ID/g) post-injection and decreased by about 40% at 24 hours (3.23±1.06 %ID/g). Kidney uptake was highest at 24 hours (10.83±2.61 %ID/g) and decreased by about 40% at 120 hours (6.74±2.61 %ID/g). The blood pool showed the highest activity at 4 hours (18.05±1.58 %ID/g) and dropped rapidly by approximately 70% at 24 hours (5.68±0.58 %ID/g).

[ 89 Zr]Zr-DFO-ABDB6 imaging and biodistribution in RPMI-8226 subcutaneous xenograft models. ( A ) Micro-PET imaging of [ 89 Zr]Zr-DFO-ABDB6 in RPMI-8226 subcutaneous xenograft tumor mice at 4, 24, 72, and 120 hours p.i. ( B ) ROI analysis of [ 89 Zr]Zr-DFO-ABDB6 at different time points. ( C ) The TMR of [ 89 Zr]Zr-DFO-ABDB6 at different time points. ( D ) Biodistribution of [ 89 Zr]Zr-DFO-ABDB6 in various tissues at time points from 4 hours to 120 hours. PET, positron emission tomography; p.i., post-injection; ROI, regions of interest; SUVmax, maximum standardized uptake value; TMR, tumor/muscle ratio; %ID/g, injected dose per gram.

Biodistribution, dose escalation, and toxicity study of [ 177 Lu]Lu-DOTA-ABDB6

We conducted a biodistribution study of [ 177 Lu]Lu-DOTA-ABDB6 in non-tumor-bearing nude mice at various time points: 4 hours, 24 hours, 72 hours, and 168 hours post-injection ( online supplemental figure 3 ). Consistent with the biodistribution of [ 89 Zr]Zr-DFO-ABDB6, liver uptake of [ 177 Lu]Lu-DOTA-ABDB6 peaked at 4 hours (19.31±3.29 %ID/g) post-injection, then decreased by about 40% at 24 hours (11.90±1.16 %ID/g). In contrast, kidney uptake reached its maximum at 72 hours (28.82±2.22 %ID/g) post-injection and then declined by approximately 60% by 168 hours (11.94±2.71 %ID/g). The blood pool showed the highest activity (21.56±5.86 %ID/g) at 4 hours post-injection, followed by a rapid decrease of roughly 80% by 72 hours (5.04±0.09 %ID/g).

Next, we performed a dose escalation study to identify the safe dose range using non-tumor-bearing nude mice. The study included four dose groups (n=4 per group) receiving 1.85 MBq (50 µCi), 3.7 MBq (100 µCi), 11.1 MBq (300 µCi), and 22.2 MBq (600 µCi) of [ 177 Lu]Lu-DOTA-ABDB6, along with a control group given saline (n=4). As shown in figure 5A , doses of 22.2 MBq (600 µCi) and 3.7 MBq (100 µCi) resulted in weight loss or fluctuations during the first 2 weeks, followed by gradual weight gain. Conversely, mice in the other groups—receiving 1.85 MBq (50 µCi) or 11.1 MBq (300 µCi)—showed consistent increases in body weight. All mice survived the 4-week observation period ( figure 5B ).

[ 177 Lu]Lu-DOTA-ABDB6 dose selection study in non-tumor-bearing nude mice. ( A ) Body weight changes post [ 177 Lu]Lu-DOTA-ABDB6 treatment. ( B ) Survival of mice post-treatment.

Additional studies, including blood cell counts ( figure 6A–D ) and assessments of liver and kidney function ( figure 6E–H ), were conducted to evaluate both acute (day 7 post-treatment) and long-term (day 28 post-treatment) toxicity in mice given 1.85 MBq (50 µCi), 3.7 MBq (100 µCi), 11.1 MBq (300 µCi), and 22.2 MBq (600 µCi) of [ 177 Lu]Lu-DOTA-ABDB6. The groups receiving 300 µCi and 600 µCi showed significant reductions in white blood cell ( figure 6B ) and platelet ( figure 6C ) counts within 7 days compared with the saline group, along with notable increases in alanine aminotransferase (ALT) ( figure 6E ), aspartate aminotransferase (AST) ( figure 6F ), and creatinine (CREA) ( figure 6G ) levels. However, by 28 days, there were no significant differences in blood cell counts or liver and kidney function between the treatment groups and the saline control.

Acute and long-term toxicity assessment by blood analysis. Non-tumor-bearing nude mice received 1.85 MBq (50 µCi), 3.7 MBq (100 µCi), 11.1 MBq (300 µCi), 22.2 MBq (600 µCi) [ 177 Lu]Lu-DOTA-ABDB6 or no treatment (saline group). ( A–D ) Blood cell count analysis of mice 1 week and 4 weeks post-treatment. ( E–H ) Liver and kidney function analysis of mice 1 week and 4 weeks post-treatment. ALT, alanine aminotransferase; AST, aspartate aminotransferase; CREA, creatinine; HGB, hemoglobin; PLT, platelet; RBC, red blood cells; WBC, white blood cell.

As shown in online supplemental figures 4 and 5 , treatment-related abnormalities were found in spleen tissue sections. Lymphoid hyperplasia was observed in the spleens of mice in the 100 µCi, 300 µCi, and 600 µCi groups at both 7 ( online supplemental figure 4 ) and 28 ( online supplemental figure 5 ) days post-treatment, with the most notable hyperplasia seen in the 600 µCi group at 28 days. No significant pathological changes were noted in the heart, lungs, liver, or kidneys across any group.

Therapeutic evaluation of [ 177 Lu]Lu-DOTA-ABDB6 in RPMI-8226 xenografts

Since both 300 µCi and 600 µCi doses of [ 177 Lu]Lu-DOTA-ABDB6 cause acute and chronic toxicity in nude mice, and considering the weaker immune function of NCG mice compared with nude mice, we subsequently administered [ 177 Lu]Lu-DOTA-ABDB6 at doses of 40 µCi and 80 µCi to NCG mice bearing RPMI-8226 xenografts for the therapeutic study. Cohorts of six RPMI-8226 xenograft-bearing NCG mice were randomized to receive injections of saline, the non-labeled nanobody ABDB6, or [ 177 Lu]Lu-DOTA-ABDB6 at activity levels of 40 µCi or 80 µCi. The body weight measurements of these mice are shown in figure 7A . The results indicate that doses of 40 and 80 µCi of [ 177 Lu]Lu-DOTA-ABDB6 cause a temporary decrease in body weight, with a return to initial weights within 2 weeks after treatment. As shown in figure 7B , treatment with 40 and 80 µCi of [ 177 Lu]Lu-DOTA-ABDB6 significantly inhibited tumor growth in an activity-dependent manner, while mice in the control groups (saline and ABDB6) showed rapid tumor growth. The survival curve ( figure 7C ) shows that the median survival for mice treated with saline and ABDB6 was 36 and 35 days, respectively. In contrast, the median survival for the groups receiving 40 and 80 µCi of [ 177 Lu]Lu-DOTA-ABDB6 improved to 61 and 69 days, respectively. Overall, these findings demonstrate that doses of 40 and 80 µCi of [ 177 Lu]Lu-DOTA-ABDB6 are effective in treating RPMI-8226 xenografts. We also compared the two dosages in figure 7 . As illustrated, a dose of 80 µCi is more effective than 40 µCi at inhibiting tumor growth and prolonging survival (*p<0.05; ***p<0.001).

Therapeutic efficacy of [ 177 Lu]Lu-DOTA-ABDB6 in the RPMI-8226 xenograft-bearing NCG mice. ( A ) Normalized body weight. ( B ) Mean tumor volumes. ( C ) Survival curves. Data are presented as mean±SD (n=6), *p<0.05; ***p<0.001).

Discussion

MM is a malignancy of plasma cells and has undergone revolutionary progress in theranostics—the integration of diagnostic imaging and targeted therapy. This development has paved the way for personalized treatment strategies. Molecular imaging and radiopharmaceuticals are critical to this new approach, significantly improving detection, prognostic assessment, and therapeutic accuracy in MM. While 18 F-FDG PET/CT remains the mainstay of MM imaging for detecting medullary and extramedullary lesions and evaluating treatment responses, new tracers targeting biomarkers highly expressed in MM have also demonstrated great promise in preclinical and translational studies. For example, 68 Ga-Pentixafor targets the CXCR4 receptors, which are overexpressed in MM, and shows significant potential in identifying 18 F-FDG-negative lesions. 24 , 27 Its theranostic potential is further emphasized by its combination with 177 Lu-Pentixafor for therapeutic applications. 24 CD46-targeted probes, such as the antibody-based agent [ 89 Zr]Zr-DFO-YS5, enable immuno-PET imaging of CD46, a cancer-specific antigen in MM. Meanwhile, the therapeutic agent [ 225 Ac]Ac-DOTA-YS5 has demonstrated substantial tumor suppression in murine MM models, highlighting the effectiveness of alpha-particle therapy. 25

CD38-targeted agents, specifically radiolabeled anti-CD38 antibodies, 28 , 31 sdAb, 31 32 or peptide, 33 34 are currently under investigation for both imaging and therapeutic applications, leveraging the widespread expression of CD38 in MM cells. Moreover, novel targets such as BCMA and CD70 are also being exploited for molecular imaging of disseminated MM. 35

In this study, we successfully developed a CD70-targeted theranostic pair consisting of [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6. The former is used for immuno-PET imaging of CD70 expression and to validate the pharmacokinetics of ABDB6, while the latter is used for effective RIT of MM. From a diagnostic standpoint, same-day molecular imaging approaches using [ 68 Ga]Ga-NOTA-RCCB6 and [ 18 F]RCCB6 may be more practical for routine clinical application. 19 20 However, the single-valent sdAb RCCB6 cannot be used for RIT due to its short circulation time and high renal accumulation. Thus, we engineered ABDB6 to possess albumin hitchhiking properties, 15 aiming to prolong its circulation time and reduce its accumulation in the kidneys. RCCB6 is a nanobody with a molecular weight of approximately 15 kDa, while ABDB6 is an engineered derivative formed by fusing RCCB6 with the albumin-binding agent ABD035, which increases its molecular weight to about 25 kDa. 15 This modification prolongs its retention time in the bloodstream and enhances tumor uptake; however, it also increases hematological toxicity. These differences underscore the characteristics of the precursor compounds. Due to its shorter half-life, 18 F is suitable for labeling smaller precursor molecules, making [ 18 F]RCCB6 appropriate for screening CD70-positive clinical patients. In contrast, 89 Zr, with its longer half-life, is more suitable for labeling larger precursor compounds. [ 89 Zr]Zr-DFO-ABDB6 is primarily used to observe the distribution of ABDB6 in major organs over a 1-week period through imaging and to assess the affinity of ABDB6 for CD70, thereby predicting the side effects and efficacy of [ 177 Lu]Lu-DOTA-ABDB6. 177 Lu is a widely used beta-ray therapeutic isotope characterized by relatively low emission energy (maximum Eβ of 0.49 MeV) and limited emission range (maximum 1.6 mm). The radionuclide 177 Lu can directly kill tumor cells by emitting high-energy beta rays. Therefore, it has been widely used to create effective radiopharmaceuticals with promising potential for RIT. 22 23 In recent years, the clinical use of theranostics has emerged, with Food and Drug Administration (FDA) approval of two 177 Lu-labeled peptide drugs: [ 177 Lu]Lu-PSMA-617 (Pluvicto), targeting prostate-specific membrane antigen (PSMA) for prostate cancer, 36 and [ 177 Lu]Lu-DOTA-TATE (Lutathera), targeting somatostatin receptors for neuroendocrine tumors. 37 Clinical trials of FDA-approved and emerging radiopharmaceuticals are continually emerging. The field of theranostics presents significant potential for emerging clinical applications. Immuno-PET/CT imaging with [ 89 Zr]Zr-DFO-ABDB6, which targets CD70, may further support the potential clinical application of [ 177 Lu]Lu-DOTA-ABDB6 in the treatment of MM.

The preliminary therapeutic experiments showed that a single high-dose administration of [ 177 Lu]Lu-DOTA-ABDB6 (300 and 600 µCi) caused a decrease in white blood cell and platelet counts, along with an increase in ALT, AST, and CREA levels during the first week. These parameters returned to normal within 4 weeks of administration, indicating short-term hematological toxicity at these doses. Therefore, for subsequent treatment experiments using the subcutaneous RPMI-8226 model, we selected two lower-dose regimens (40 and 80 µCi). Both doses were generally well tolerated, as shown by the stable body weight during treatment. Treatment with either 40 µCi or 80 µCi of [ 177 Lu]Lu-DOTA-ABDB6 effectively inhibited tumor growth within the first 40 days and significantly extended survival in tumor-bearing mice. The 80 µCi dose demonstrated more promising treatment efficacy. However, tumor relapse was observed in both groups during later stages of the study, indicating the need for repeated treatment. Because the tumor models were established in NCG mice, their severe immunodeficiency and weight loss prevented repeated treatments in our experiments. Although acute and chronic toxicity were observed with 300 µCi and 600 µCi doses of [ 177 Lu]Lu-DOTA-ABDB6 in non-tumor-bearing nude mice, no mouse fatalities occurred. ABDB6 was developed by conjugating the human serum albumin-binding ABD molecule to RCCB6, which increased the molecular weight by approximately 25 kDa. This modification extended the blood circulation time and enhanced tumor uptake, although at the cost of increased hematological toxicity. In our preliminary experiments, we encountered challenges in establishing multiple myeloma xenografts in nude mice. Therefore, we used severely immunocompromised NCG mice in the therapeutic experiments. Due to the inherent limitations of this severely immunodeficient mouse model, the therapeutic window was narrow. Nonetheless, [ 177 Lu]Lu-DOTA-ABDB6 treatment at a therapeutic dose of 40–80 µCi still significantly inhibited tumor growth, with no NCG mouse fatalities observed throughout the 55-day observation period. It can be inferred that in future clinical applications, patients in poorer physical condition may require lower drug doses than those in better health. However, even at reduced doses, the medication can still inhibit tumor growth and provide therapeutic benefits.

We expect that optimizing treatment protocols, such as administering repeated treatments at earlier stages or replacing 177 Lu with other alpha radionuclides, may improve treatment outcomes. Alpha radionuclide-labeled radioimmunotherapy agents could have unique value for treating minimal residual disease. 38 Despite the relatively complex procedures of pretargeted radioimmunotherapy, 38 39 it remains a promising platform for enhancing the therapeutic index and minimizing adverse toxicities. Our ongoing research continues to explore these possibilities in both subcutaneous and disseminated models of MM.

In preclinical and translational studies, RCCB6 labeled with 18 F or 68 Ga has demonstrated significant diagnostic value in nasopharyngeal carcinoma, 40 clear cell renal carcinoma, 19 20 and lymphoma. 15 19 20 40 In this study, a patients with MM underwent [ 18 F]RCCB6 immuno-PET/CT imaging, which showed that the number of lesions detected by [ 18 F]RCCB6 was approximately twice that identified with 18 F-FDG. Additionally, both the TBR and TMR of [ 18 F]RCCB6 in this patient were significantly higher than those of 18 F-FDG. These results indicate that [ 18 F]RCCB6 has better diagnostic potential for lesions related to MM. The tissue sample from this patient with MM could not be obtained, so we were unable to assess CD70 expression. Future studies should include a larger group of patients with MM to evaluate better the diagnostic performance of [ 18 F]RCCB6 in this disease.

CD70-targeted cancer theranostics and radiotheranostics are emerging approaches for managing CD70-positive human malignancies. 12 41 Our study presents a novel theranostic pair, [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6, and reports its preclinical value in the MM model. We first demonstrate the effectiveness of this drug. Future research should include additional comparisons and examine the potential advantages of combining this treatment with existing therapies to see if the combined approach is more effective than monotherapy. Clinical trials are necessary to assess the potential clinical benefit.

Methods

Cell culture and animal studies

In this study, the RPMI-8226 and U266 cell lines were kindly provided by Dr Yunfeng Fu from the Third Xiangya Hospital of Central South University. Meanwhile, Professor Jia Feng from Peking University Shenzhen Hospital supplied the MM1.S cell line. The cells were cultured in Roswell Park Memorial Institute -1640 (RPMI-1640) medium supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin, and maintained at 37°C in a 5% CO 2 atmosphere.

Approval for animal experiments (Approval Number: 2023–631, 2022-579) was granted by the Animal Care and Use Committee at ShenZhen PKU-HKUST Medical Center. The mice were housed in a pathogen-free environment, with unrestricted access to food and water, and kept on a 12-hour light/dark cycle. Female NCG mice, aged 4–6 weeks and derived from GemPharmatech, received subcutaneous injections of 5×10 6  MM tumor cells into their flanks.

Immunohistochemical staining for CD70

Human MM tumor specimens obtained from Peking University Shenzhen Hospital (Approval Number: 2024–132), along with tumors derived from MM models, were stained using an anti-CD70 antibody (ab300083; Abcam). To inhibit endogenous peroxidase activity, the slides underwent three microwave treatment sessions, each lasting 5–10 min, followed by a 15-minute incubation with 3% hydrogen peroxide at room temperature. The primary antibody was diluted to 1:500 and incubated overnight at 4°C. Subsequently, a secondary antibody, horseradish peroxidase (HRP)-conjugated and specific to rabbit, was applied at a dilution of 1:200 for 50 min at room temperature. Visualization was performed using 3,3′-diaminobenzidine for 5 min, and the sections were counterstained with Mayer’s hematoxylin for 15 min, then rinsed in distilled water.

Clinical [ 18 F]RCCB6 immuno-PET/CT in patients with MM

The clinical study was approved by the Medical Ethics Committee at Peking University Shenzhen Hospital and registered with the Chinese Clinical Trial Registry (ChiCTR) under the ID ChiCTR2400091247. Informed consent was obtained from each patient before participation. One male patient, 61 years old, with pathologically confirmed MM, was included. The dose of [ 18 F]RCCB6 administered was 226.4 MBq. The patient underwent PET/CT imaging 1 hour after injection. Before receiving [ 18 F]RCCB6 immuno-PET/CT imaging, the patient had 18 F-FDG PET/CT to identify lesions. Imaging was performed using a GE HealthCare PET/CT scanner, and the SUVmax was measured using a standardized method.

Chemistry and radiolabeling

In this study, the CD70-targeting nanobody ABDB6 (Biointron Biological) was labeled with 89 Zr, 18 F, and 177 Lu, primarily based on our previous research. 42 43

DFO conjugation and 89 Zr-labeling of ABDB6

Dissolve 2–4 mg of ABDB6 in PBS and adjust the pH of the solution to 8.9–9.1 using a 0.1 M Na 2 CO 3 buffer (pH=11.4). Subsequently, dissolve freshly prepared DFO (1222468–90-7; Macrocyclics) in dimethyl sulfoxide (DMSO). Add the DFO solution to the protein solution at a DFO-to-ABDB6 molar ratio of 5:1. Gently shake the reaction mixture at room temperature for 45 min, then purify the conjugated protein using a PD-10 desalting column (GE HealthCare) with PBS as the mobile phase. Finally, concentrate the conjugated protein solution with an ultrafiltration tube (10 kDa, Merck). 89 Zr was freshly obtained from Guangdong Sinotau Molecular Imaging Technology. For the [ 89 Zr]Zr-labeling process, a solution of [ 89 Zr]Zr-oxalic acid (20 µL, 74 MBq) was mixed with 2–3 µL of 1 M Na 2 CO 3 buffer solution, adjusting the pH to 7.1–7.3. Subsequently, 200 µL of DFO-ABDB6 (200–300 µg) was added to the reaction vial. The mixture was incubated for 1 hour at room temperature with gentle shaking. The final radiopharmaceuticals were purified on PD-10 columns, and radiochemical purity (RCP) was assessed by thin-layer chromatography.

DOTA conjugation and 177 Lu-labeling of ABDB6

Subsequently, 6 mg of ABDB6 solution in PBS was added to the PD-10 column, and the volume was adjusted to 2.5 mL with the same carbonate buffer solution (pH=9). An additional 2.5 mL of the solution was eluted and collected using the 0.05 M NaHCO 3 solution. A precise amount of p -SCN-Bn-DOTA (1 27 985–74-4; Macrocyclics) was weighed to ensure a molar ratio of p -SCN-Bn-DOTA to ABDB6 of 10:1. The p -SCN-Bn-DOTA was subsequently combined with the ABDB6 solution and thoroughly mixed. The resulting reaction mixture was placed on a shaker at room temperature for 1.5 hours, and the conjugated protein was then purified using a PD-10 column with PBS. [ 177 Lu]LuCl 3 was freshly obtained from Guangdong Sinotau Molecular Imaging Technology. For the [ 177 Lu]Lu-labeling process, a solution of [ 177 Lu]LuCl 3 (59.2 MBq) was mixed with 200 µL of 0.25 M NaOAc (pH=5.5). Subsequently, 200 µL of DOTA-ABDB6 (200–300 µg) was added to the reaction vial. The mixture was incubated for 1 hour at 42°C with gentle shaking. The final radiopharmaceuticals were purified using PD-10 columns, and the RCP was assessed using iTLC. The in vitro stability of [ 89 Zr]Zr-DFO-ABDB6 and [ 177 Lu]Lu-DOTA-ABDB6 was evaluated by incubating them in PBS at 4, 12, 24, and 48 hours.

In vivo imaging studies

PET imaging using [ 89 Zr]Zr-DFO-ABDB6 was conducted on RPMI-8226 tumor-bearing mice. A total of 3.7 MBq of [ 89 Zr]Zr-DFO-ABDB6 was intravenously administered to the tumor-bearing mice (n=3 for each group), followed by small-animal PET scans (wBIT, Shenzhen Bay Lab and Bay Imaging Technology) at various time points. The PET images were reconstructed using three-dimensional ordered subset expectation maximization, and regions of interest (ROIs) were manually delineated using PMOD software (PMOD Technologies, Zurich, Switzerland). Subsequently, the ROI data were normalized to the administered activity, allowing for the parameterization of images as a percentage of the injected dose per gram (%ID/g).

Biodistribution study

RPMI-8226 tumor-bearing mice (n=3) were administered 0.74 MBq of [ 89 Zr]Zr-DFO-ABDB6 or 0.74 MBq of [ 177 Lu]Lu-DOTA-ABDB6. Subsequently, major tissues including blood, liver, heart, kidney, lung, spleen, stomach, small intestine, large intestine, pancreas, muscle, bone, and tumor were harvested at various post-injection time points. The collected samples were weighed and analyzed using a γ-counter, with the data expressed as %ID/g.

Dose study and toxicity analysis with [ 177 Lu]Lu-DOTA-ABDB6

To determine the safe dose for the therapeutic study, a dose escalation study was conducted using Balb/c-Nude female mice (GemPharmatech) without tumors to assess toxicity. Groups of nude mice (n=4 per group) received doses of [ 177 Lu]Lu-DOTA-ABDB6 at 0, 1.85 MBq, 3.7 MBq, 11.1 MBq, and 22.2 MBq (equivalent to 0, 50, 100, 300, and 600 µCi) per mouse. The health status, signs of stress, and body weight of the mice were monitored every 2–3 days. Mice were sacrificed if they lost more than 20% of their body weight or showed visible signs of distress.

To evaluate the potential toxicity of [ 177 Lu]Lu-DOTA-ABDB6, we performed assessments including blood count, liver and kidney function tests, and histological staining. Blood and serum samples were analyzed using an automatic biochemical analyzer (Rayto Biochemical Analyzer). The blood cell count included white blood cells, red blood cells, platelets, and hemoglobin levels. Liver function was assessed by measuring AST and ALT. Kidney function was evaluated by analyzing uric acid and creatinine levels. Histological analysis mainly involved H&E staining on the heart, liver, spleen, lung, and kidney.

[ 177 Lu]Lu-DOTA-ABDB6 therapy studies in the RPMI-8226 xenograft model

In this therapeutic study, NCG mice with RPMI-8226 tumors approximately 64 mm 3 in volume were given a single intraperitoneal dose of [ 177 Lu]Lu-DOTA-ABDB6 at either 40 µCi (1.48 MBq) (n=6) or 80 µCi (2.96 MBq) (n=6). A control group received either 13 µg of unlabeled ABDB6 per mouse (n=5) or an equivalent volume of saline solution (n=6). Tumor dimensions were measured two times per day with a digital caliper, and tumor volumes were calculated using the modified ellipsoid formula: volume=0.5×length×width². Animal survival data were analyzed with Kaplan-Meier analysis (GraphPad Prism). Body weight was recorded every 2 days. Mice were euthanized if any of the following conditions were met: tumor volume reached 1,800 mm³, body weight loss exceeded 20%, or visible signs of distress were observed.

Statistical analysis

Data analysis was performed using GraphPad Prism (V.9), which facilitated determination of statistical significance (p<0.05) and graph preparation.

Supplementary material

Acknowledgements

We thank our hospital colleagues for their helpful discussions in carrying out the studies.

Footnotes

Funding: This research was funded by grants from the Science, Technology and Innovation Commission of Shenzhen Municipality (Grant numbers: JCYJ20220530160213030 to MD, JCYJ20220531094206013 to JL, JCYJ20250604183821029 to PC), Peking University Shenzhen Hospital Scientific Research Initiation Fund Project (Grant number: KYQD2024368 to MD), Guangdong Basic, Sanming Project of Medicine in Shenzhen（Grant number: SZSM202211037), Applied Basic Research Foundation (Grant number: 2025A1515010950 to MD), National Natural Science Foundation of China (Grant number: 82327809 to QP), National Key Research and Development Program of China (Grant numbers: 2024YFF0727400 to SX, 2024YFF0507700 to QP).

Provenance and peer review: Not commissioned; externally peer reviewed.

Patient consent for publication: Not applicable.

Ethics approval: This study involves human participants and was approved by Institutional Ethic Reviewing Board of Peking University Shenzhen Hospital (2024-132). Participants gave informed consent to participate in the study before taking part.

Data availability free text: Data are available upon reasonable request. Additional data are available from WW and MD upon reasonable request.

Data availability statement

Data are available upon reasonable request.

References

Associated Data

Supplementary Materials

Data Availability Statement

Data are available upon reasonable request.