Background
Prostate cancer (PCa) is the most commonly diagnosed cancer in men, and the second most lethal cancer among North American men [
1]. Localized PCa is treated with surgery and/or radiation therapy. For those patients who have advanced androgen sensitive disease, the standard treatment is usually different forms of androgen deprivation therapy [
2]. If the disease progresses to castration-resistant prostate cancer (CRPC), a state of resistance to the therapeutic regiment of serum androgen suppression, the prognosis becomes poor, with an expected survival for patients with metastases of less than 19 months [
3]. Resistance to androgen suppression is complex and relies on, e.g., changes in androgen receptor (AR) sensitivity, AR specificity, AR ligand independence, or bypassing of the AR pathway [
4]. Commonly, CRPC is treated with the continuation of androgen deprivation, chemotherapy, and external beam radiation therapy (EBRT). However, there are also several new approaches with promising results for combating CRPC, ranging from anti-androgen synthesis therapy to immunotherapy [
5]. Radionuclide therapy with the α-emitting bone seeker
223Ra (Xofigo, Bayer) has shown encouraging results with an increase of median overall survival of 3.6 months [
6]. Also, prostate-specific membrane antigen (PSMA)-targeted radioimmunotherapy (RIT) has been extensively studied.
177Lu-labeled J591 targeting the external domain of PSMA has shown efficacy in phase II clinical trials [
7], which is encouraging for the development of novel RIT approaches in CRPC treatments. However, today, the development of resistance to the therapeutic regimens eventually renders all therapies suboptimal.
RIT is one of few therapeutic options, where the outcome and tolerable dose can be better predicted for each individual treatment based on dosimetry [
8]. This is possible since the absorbed dose to both normal organs and tumor can be correlated to the observed biokinetics. Usually, medium energy β
−-emitters are the choice of therapeutic radionuclides. The resulting crossfire effect circumvents any need for targeting all malignant cells and reduces the impact of low tumor penetration and heterogeneous antigen localization and therefore gives better tumor response [
9]. Even though RIT of solid cancers is not common and could still be considered to be in evaluation [
10], a RIT rationale against PCa could work since PCa is a relatively radiosensitive disease and since metastatic PCa localizes to tissue that receive high levels of circulating antibodies, such as bone marrow and lymph node metastases [
11].
Human kallikrein-related peptidase 2 (hK2) is a prostate-specific serine protease normally highly specific to the prostate [
12] with low or no expression in other organs. hK2 shares about 80 % homology with prostate-specific antigen (PSA) and is encoded by the human kallikrein 2 gene (KLK2) [
13,
14]. KLK2 is an AR-regulated gene, and since AR signaling is in general retained in metastasized PCa and CRPC [
15,
16], targeting a protein downstream of the AR, like hK2, could be beneficial. Data from patient tissues has suggested that the relative expression of KLK2 is increased in malignant tissue compared to benign and healthy prostatic tissue, and the intensity of hK2 immunostainings in, e.g., lymph node metastasis correlates better with PCa tumor grade than that of PSA [
13,
14].
Our group recently published a work on a new radioimmunoconjugate based on the
111In-labeled murine monoclonal antibody 11B6 (m11B6) suitable for molecular imaging of free, not associated with protease inhibitors, hK2 in PCa [
17]. 11B6 is an IgG
1 with a high affinity and specificity for free hK2 [
18].
111In-labeled m11B6 showed high tumor accumulation in both subcutaneous (s.c.) as well as bone-enclosed PCa xenografts, mimicking metastasized PCa [
17]. This could translate to high absorbed doses to the tumor using a β-emitter like
177Lu. Additionally,
177Lu has a gamma component which allows the in vivo distribution of this radionuclide to be followed using single photon emission computed tomography (SPECT).
Here, we present results on the therapeutic efficacy and the biokinetics of m11B6 labeled with 177Lu in a pre-clinical setting. This is, to our knowledge, the first data on RIT targeting a secreted antigen belonging to the kallikrein-related peptidase family. Additionally, we evaluated therapy planning with a dedicated dosimetry model in an attempt to define the pre-clinical therapeutic window of 177Lu-labeled m11B6.
Discussion
In this study, we have demonstrated the therapeutic efficacy of
177Lu-m11B6, a hK2-targeting radioimmunoconjugate, in prostate cancer xenografts.
177Lu-m11B6 displays interesting therapeutic properties and high uptake in subcutaneous LNCaP xenografts. For the lowest administrated activity of
177Lu-m11B6 of 10 MBq, a median survival of 88 days was achieved compared to 39 days median survival on average for the control groups (Fig.
5a). For administrated activities of
177Lu-m11B6 in the range of 19 and 36 MBq, the survival was 100 % up to 120 days; a high therapeutic effect attributed to the high estimated absorbed doses delivered to the tumor of 48, 92, and 180 Gy (Table
2). This is in the range of what creates clinical response in a solid tumor, like PCa, using EBRT [
28]. In the majority of RIT studies conducted in solid tumors, absorbed doses below 50 Gy are reported [
29]; hence, our data for
177Lu-m11B6 is promising. Note however that these comparisons should be interpreted with care, due to the difference in modality and subsequent difference in the dose rates given. Still, the lower absorbed dose, 48 Gy, was enough to allow for efficient treatment and with less concern for bone marrow toxicity (4.5 Gy versus 8.6 Gy and 16 Gy, see Table
2).
A potential threat to the use of m11B6 in RIT is that the organ which received the highest absorbed dose was the submandibular glands; however, the uptake was not significantly reduced by pre-dosing (Fig.
2c) which means that it might not be specific. One explanation could be that the high uptake in the submandibular glands of the murine model could be due to cross-reactivity. Murine submandibular glands are known to be abundant with mice kallikreins; however, the KLK2 gene is a silenced pseudogene in mice [
30]. In humans, salivary gland tissue immunostaining towards hK2, though positive, has been shown to be considerably less intense than that of the prostate and prostate carcinoma [
31].
The maximum tumor uptake of
177Lu-m11B6 was higher than previously seen with
111In-m11B6, probably due to the use of a higher antibody dose (30 versus 20 μg) [
17]. A higher xenograft uptake could in theory contribute to less toxicity. However, the calculated absorbed dose to the bone marrow is similar for both cases. Additionally, several normal organs have theoretically higher absorbed doses with
177Lu-m11B6-based dosimetry calculations compared to that with
111In-m11B6 (Table
1). Still, even at 16 Gy to the bone marrow, a 4-Gy higher absorbed dose than the predicted tolerable dose, no adverse effects were observed. These findings combined highlight not only the difficulties of predicting toxicity but also the potential of pre-therapy planning, with, e.g.,
111In labeled immunoconjugates, for finding reasonable ranges of therapeutic activities to be administered in pre-clinical studies. Pre-therapy dose planning, as done in clinical studies, could maybe replace the maximum tolerated dose or the maximum tolerated activity studies in pre-clinical therapy studies.
RIT in general has so far not been very successful in solid tumors due to the limited penetration of antibodies. RIT with PSMA targeting has however shown promise in phase II clinical trials [
7]. This PSMA is not only confined to the cells of the prostate but is also expressed in other organs such as the duodenum, the kidneys, the brain, and, like hK2, the salivary glands [
32]. Small molecules, labeled with therapeutic radionuclides, targeting PSMA have been developed, e.g., in a recent study, it shown to be effective with limited toxicity to normal organs, including the salivary glands [
33]. In the specific case of hK2 targeting, it is not clear how the size of the antibody and the presence of a Fc region affects the uptake and biokinetics of the radioimmunoconjugate. This would have to be further investigated before making any decisions on whether to proceed with testing smaller proteins or other small molecules. Even though smaller molecules could increase the tumor penetration and lower the absorbed dose to normal organs due to faster clearance, the longer retention time of a full antibody contributes to a higher tumor uptake. Schmidt et al. used theoretical analysis to predict tumor uptake in relation to size and affinity [
34]. Their model showed that tracers with the smallest and largest molecular masses exhibited the highest tumor uptake, whereas tracers with intermediate mass (25–60 kDa) displayed the lowest tumor uptake. It has been shown that high-affinity radioimmunoconjugates (low
K
D values) generally have low penetration and reduced diffusion due to the so-called binding site barrier [
35], known to limit therapeutic efficacy. An affinity of 1 nM has been considered to be optimal in this type of therapeutic setting [
36], but m11B6, being a high-affinity antibody (
K
D of 0.65 pM), has here shown high tumor uptake and good therapeutic efficacy. For the specific case of shedded antigens and high-affinity binders, Pak et al. has suggested that this combination might circumvent the binding site barrier issues seen with high-affinity binders [
37]. The fact that hK2 is a secreted antigen [
38] could therefore potentially increase its availability for targeting and, in combination with the high affinity of
177Lu-m11B6, lead to better tumor penetration.
hK2 can be found in the serum of patients, and the serum levels of hK2 have been found to vary with degree of disease, e.g., Stephan et al. found that grade G3 cancers with a Gleason score larger than 7 had a hK2 median of 0.23 μg/L [
39]. Another study by Steuber et al. reported values roughly half of that for patients with extracapsular extension or seminal vesicle invasion [
40]. It is not clear how serum levels, albeit low, of hK2 would affect RIT towards the antigen. However, these levels can be continuously monitored.
Due to the problem with resistance, against chemotherapy and targeted therapies, and the complex biology behind advanced PCa, the only reasonable approach is most likely a multi-combinatorial therapy rationale, an approach that comes with the necessity of expanding the clinical therapy options. As a novel therapy target, hK2 has several possible advantages: it is highly restricted, although not completely, to prostatic tissue and it has a possibly higher expression in some patients following EBRT [
41] and a seemingly retained expression in the different clinical phases of PCa. It is therefore plausible to assume that the use of hK2 targeting therapy might not be limited to only one stage of the disease. And, even though it has not yet been investigated in a model of castration-resistant disease, it could potentially be useful in CPRC patients as well as in patients undergoing treatment for disseminated androgen responsive disease.
More studies are needed in order to further investigate the underlying factors and mechanisms behind the therapeutic effects of 177Lu-m11B6, but the current results present a radioimmunoconjugate with high potential for the treatment of prostate cancer. Remarkably, an administrated activity of 10 MBq showed a survival of 88 days, and a twofold increase of the administered activity gave a 100 % survival in this group up to 120 days. For future intended use in humans, we are now working on a new less immunogenic humanized version of the radioimmunoconjugate.
Acknowledgements
We wish to thank Anna Åkesson and Susanne Strömblad at the Medical Radiation Physics; Gustav Grafström at Lund University Bioimaging Center (LBIC) for the technical assistance. LBIC, Lund University is gratefully acknowledged for providing the experimental resources. Oskar Vilhelmsson Timmermand was supported by The Research School in Pharmaceutical Sciences (FLÄK, Lund University), and David Ulmert by the David H Koch Young Investigator Award from the Prostate Cancer Foundation. In addition, this study was performed with generous support from the Swedish Cancer Foundation, Mrs. Berta Kamprad’s Foundation, Gunnar Nilsson’s Foundation, Percy Falk’s Foundation, and Government funding of clinical research within the Swedish NHS (National Health Service) Lund University, Sweden (ALF).
Competing interests
Prof. Sven-Erik Strand, PhD and Dr. David Ulmert, MD, PhD are shareholders of DiaProst (Lund, Sweden) who hold patents for hK2 targeting. Dr. Thuy A. Tran, PhD holds stock options in DiaProst. Dr. Sven-Erik Strand, Dr. David Ulmert, and Dr. Thuy A. Tran are authors and co-authors of patents regarding hK2 targeting in RIT and in diagnostics. Dr. Oskar Vilhelmsson Timmermand is the co-author of a patent related to hK2 targeting in RIT. DiaProst has not financed any part of this study.
Authors’ contributions
OVT carried out the study, performed the analysis, and drafted the manuscript. TT carried out the radiochemistry and prepared the manuscript. DU initiated and participated in the planning of the study. SES participated in the design and coordination and helped to draft the manuscript. EL calculated the absorbed doses and helped draft the manuscript. All authors read and approved the final manuscript.