Abstract
Background/Aim: The current systematic review aimed to collect and analyze all available published and unpublished cases in which prostate-specific membrane antigen (PSMA)-targeted radioligand therapy (177Lu-PSMA) was used to treat non-prostatic cancer. Materials and Methods: Literature search and evidence acquisition through contacts with organizations that use 177Lu-PSMA were employed. PubMed/Medline, SCOPUS, and ScienceDirect searches were performed following PRISMA recommendations. The search strategy was to screen all articles describing 177Lu-PSMA radioligand therapy published to date with the key word “177Lu-PSMA”. These articles were collected and screened for non-prostatic cancer cases. Quality assessment was performed using the GRADE criteria. Results: A total of 713 articles were screened, and the search revealed 15 eligible records. Forty patients with a mean age of 51.2±18.5 years were treated with 177Lu-PSMA for non-prostatic cancer. Of them, 30 cases were published, and 10 were found in medical institution records. Cancers of the salivary glands were most often targeted (13/40), followed by various brain cancer types (8/40), and osteosarcoma (6/40). The authors used previously established protocols for castration-resistant prostate cancer with the dose per cycle as 6.0-7.4 GBq and the number of cycles between one and four. Toxicity was estimated as low, and 21 out of 28 patients with reported outcomes survived to the time of the publication. Conclusion: PSMA-targeted radioligand therapy was infrequently used to treat different non-prostatic cancer types in various target organs. These pioneering efforts indicate that 177Lu-PSMA can be used to treat non-prostatic cancer with PSMA expression. The toxicity of such treatment was low, and the outcome was relatively good.
The role of prostate-specific membrane antigen (PSMA) theranostics is well-known regarding prostate cancer. Non-prostatic diseases exhibiting PSMA uptake on positron-emission tomography (PET) are becoming more common as the number of scans performed increases. Recognizing those tumors exhibiting PSMA expression is essential since the theranostics approach might also be applied for them. The topic of this systematic review was developed in four phases.
Phase 1: PSMA was discovered. This 750-residue integral membrane glycoprotein known as glutamate carboxypeptidase II, which acts as a transmembrane zinc metalloenzyme, was purified in 1979, received its serological definition in 1981-82, and was immunohistochemically localized in 1983 (1-4). Finally, its role was established as “a new antigenic marker in epithelial prostatic cells and serum of prostatic cancer patients” in 1987, and it became widely known as PSMA (5).
Phase 2 began in 1999 and ended about 2008 (6-11). During these years, researchers realized that PSMA is not ‘prostate-specific’. PSMA was found in the epithelium of the prostate, urinary bladder, kidneys, liver, esophagus, stomach, small intestine, colon, testicular seminiferous tubules, hippocampal neurons and astrocytes, ependyma, and the adrenal gland. Moreover, PSMA was found in females in the ovary stroma, breast, and fallopian tubes (11). Among specific benign tissues, PSMA was detected in prostate secretory-acinar epithelium, duodenal columnar epithelium, proximal renal tubular epithelium, colonic ganglion cells, and breast epithelium (6). Among specific tumors besides prostatic, PSMA presence was detected in renal cells, transitional cells of the urinary bladder, testicular embryonal, neuroendocrine, breast, and pancreatic duct carcinomas, glioblastoma multiforme, malignant melanoma, soft tissue sarcoma, colonic adenocarcinoma, and osteosarcoma (6, 7, 9, 12). In addition to PSMA, prostate-specific antigen (PSA) was also found in different tissues in men and women (13-15). PSA is secreted into the blood, and its serum concentration is easy-detectable. That is why PSA produced by epithelial prostatic cells was suggested as a possible biomarker in non-prostatic cancers (16).
Phase 3: The most effective non-surgical treatment modality for prostate cancer (prostatic carcinoma) was established. Such intervention was needed for castration-resistant prostate cancer and metastatic castration-resistant prostate cancer. PSMA-targeted radioligand therapy was suggested for this purpose. β-Particle-emitting lutetium-177-labeled PSMA inhibitor (177Lu-PSMA) with its variations 177Lu-PSMA-617 (vipivotide tetraxetan), 177Lu-PSMA-I&T (zadavotide guraxetan), and some others were offered as a new molecular nuclear radiotherapy agent. It was tested in vivo in 2011 (17) and adopted in 2013 (18). After that, seven systematic reviews and meta-analyses, all dedicated to the role of 177Lu-PSMA in the treatment of castration-resistant prostate cancer, were published from 2016 to 2021 (19-25). In summary, 177Lu-PSMA radioligand therapy appeared to be relatively effective even for patients with metastases, prolonged overall survival and progression-free survival periods and demonstrated a low toxicity rate.
Phase 4 is the topic of the current review. It was assumed that if 177Lu-PSMA radioligand therapy was effective in treating prostate cancer, it might be equally effective in the treatment of other cancers in which PSMA is also expressed. It was hypothesized that soft-tissue and bone cancers with strong neovascular PSMA expression may be treated with PSMA-targeted radioligand therapy (26).
Following PICOS (patients, intervention, comparison, outcomes, study design) criteria (27), the current systematic review aimed to answer the following questions: How many patients, if any, with non-prostatic cancer have been treated with PSMA-targeted radioligand therapy to date? How was PSMA expression in non-prostatic tumors detected and evaluated? What types of tumors were involved, and what organs were targeted? What treatment protocols were used (dose, number of cycles, and duration of the treatment)? What results were achieved regarding treatment response, outcome, and treatment-toxicity-related complications?
Materials and Methods
Search strategy. The Authors aimed to collect as many cases as possible, published and unpublished, in which 177Lu-PSMA radioligand therapy was applied to non-prostatic cancer. Therefore, we employed literature search, conference presentations search, and evidence acquisition through contacts with organizations (hospitals, medical centers, and theranostic centers) which were using 177Lu-PSMA radioligand therapy.
Literature search and selection criteria. The current review applied recommendations presented by the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) (28). We performed PubMed/Medline, SCOPUS, and ScienceDirect searches, including publications in English and other languages, and with no restrictions on the years of publication. The search strategy was relatively simple because not many articles describing 177Lu-PSMA radioligand therapy have been published to date, and we only used “177Lu-PSMA” as a key word. All these articles were collected. The report exclusion procedure was done in several steps. First, all publications regarding prostate cancer were excluded. Secondly, all review and meta-analysis articles were excluded. Finally, all animal studies and articles describing 177Lu-PSMA technicalities without involvement of patients were also excluded. The remaining publications were investigated for 177Lu-PSMA radioligand therapy for non-prostatic cancer. Meta-analysis was not performed due to the level of evidence (IV-V) of the included studies, as determined following the 2011 Oxford Centre for Evidence-Based Medicine guidelines (29). Two reviewers searched for publications up to 1 January 2024.
We planned to assess the treatment response as a binary outcome based on serum PSA decline of ≥50% as the primary variable and any level of PSMA decrease/increase as additional binary outcomes if these variables were reported in the studies. In some cases, additional information was requested from the corresponding authors of the selected publications.
Radioligand therapy-related toxicities were assessed with the Common Terminology Criteria for Adverse Events (CTCAE) grading system (30) and divided into two groups: Low-grade toxicities (CTCAE 1/2) and high-grade toxicities (CTCAE 3/4) (Grade 5: death). If the analyzed publications did not report toxicities with the CTCAE grading system, we estimated the CTCAE grade using clinical information, the description of the complication, and other related information provided in the publication. The term ‘not reported’ was used if the toxicity information was not provided. A comparison of 177Lu-PSMA-I&T and 177Lu-PSMA-617 was not performed because it was already demonstrated that these two variations do not statistically significantly differ when applied to therapy (20).
Assessment of risk bias and quality appraisal. The selected studies did not describe prospective studies and randomized trials, and the risk of bias assessment was performed with limitations applicable to case reports. Two independent investigators (HG and OV) used the criteria of the GRADE guidelines to assess the individual quality of selected studies that involved a high-moderate-low-very low definition scale (31). All discrepancies were resolved by consensus with KKT serving as a judge. The quality appraisal of the relevant articles was performed and is reported. Assessment of risk bias was performed with the help of the Cochrane classification (32). Levels of evidence were assessed according to the Oxford Centre for Evidence-Based Medicine guidelines (29).
Additional evidence acquisition. Conference proceedings were searched for oral or poster presentations reporting 177Lu-PSMA therapy of non-prostatic tumors which were not further published in academic articles. In addition, the Authors personally contacted known medical organizations in which 177Lu-PSMA was used. While all of them were involved in treating castration-resistant prostate cancer, the inquiry concerned 177Lu-PSMA application to non-prostatic cases. The institutions were asked to provide the information to cover the above-mentioned research questions: Non-prostatic tumor 177Lu-PSMA therapy – yes/no; how many patients (age, sex); pretreatment diagnostics (PSMA expression in a given tumor case; serum PSA in ng/ml); types of tumors treated and target organs; the treatment protocol (dose, cycles, and duration); the treatment response (serum PSA decline of ≥50% – yes/no); and follow-up (outcome, survival, and toxicity-related complications according to CTCAE, version 5).
Results
The main findings, quality appraisal, and bias. We screened 713 articles, and the literature search gave 15 records that met our eligibility criteria. Figure 1 shows the PRISMA flow diagram. Table I shows a summary of the studies and unpublished cases included in the review. The quality appraisal of the relevant articles is presented in Table II. The first case of PSMA-targeted radioligand therapy for non-prostatic cancer was reported in 2017 (33) but was duplicated in a later publication. These 15 publications described 30 patients (34-48). Four were described twice, and one duplicate was excluded (33). We added 10 unpublished cases which were obtained directly from medical institutions to the review. The examined studies spanned the period from 2017 to 2023 and were carried out in Europe and Asia. Researchers from the USA, the UK, Canada, and Australia apparently did not participate in these pioneering efforts.
PRISMA flow diagram of the current study.
Summary of the studies and unpublished cases included in the review. The studies are in chronological order.
Systematic review of the first 40 cases of the 177Lu-PSMA therapy in the treatment of non-prostatic cancer. Studies included with quality appraisal.
The absolute majority of the articles presented case reports or case series (13/15; 19 patients) and were classified as level V (cases) or IV (case series) evidence. There were two retrospective studies (11 patients), and no prospective study was retrieved. The overall quality of the selected reports included was moderate to low with demographic data (83.5%), types and stages of treated tumors (100.0%), 177Lu-PSMA therapy protocol (100.0%), toxicity estimations (83.5%), and outcome (50.0%) reported. For case reports or retrospective studies, the quality of studies was graded as “low” as per the GRADE criteria (31). Most of the studies were case reports that described “first-time” events, and the selection bias was minimized. Moreover, the issue of consecutive vs. randomization was unimportant because two retrospective studies reported consecutive series of patients. However, outcomes were not reported for 15 patients, and risk of bias due to missing outcome data was estimated as serious according to Cochrane classification (32). The quality of unpublished cases and conference presentations was high because they were selected on the basis of all the above-mentioned variables and their presentation was supervised by us. Evaluating the contribution of possible covariates in the heterogeneity was impossible because the selected articles indicated 16 different types of tumors in 11 organs (Table I). Having 40 cases with 16 different tumor types made a sound statistical analysis impossible.
Specific results. Answering our five review questions, we received the following results:
A total of 40 patients within the age range from 19 to 74 years (mean 51.2±18.5 years), 24 males, 12 females, and four with unreported sex, were reported to have been treated with 177Lu-PSMA for non-prostatic cancer. Of them, 30 cases were published, and 10 were found in medical institution records or were presented at conferences (Table I).
PSMA expression in non-prostatic tumors was reported as immunohistochemical expression in all studies selected, either as a statement of the fact (14 publications) or with an indication of the expression percentage (one publication) and ranged from 5% to 95%. Eight unpublished cases presented PSMA expression as immunohistochemical expression, which ranged from 5% to 50%, and nine unpublished cases presented serum PSA concentration, which ranged from 1.1 to 23.2 ng/ml.
Despite the small number of cases, the reports indicated numerous target organs and types of cancer (Table I). Cancers including acinic cell carcinoma, adenoid cystic carcinoma, glioblastoma multiforme, hepatocellular carcinoma, mutant astrocytoma, metastatic testicular mixed germ-cell tumor, oral squamous cell carcinoma, osteosarcoma, papillary thyroid carcinoma, radioiodine-refractory differentiated thyroid cancer, melanoma, and wild-type glioblastoma were targeted. The breast, bones, kidneys, oral cavity, salivary glands, tongue, skin, thyroid gland, brain, testicles, and the liver were involved. Various cancers of the salivary glands were most often targeted (13/40), followed by various brain cancer types (8/40), and osteosarcoma of long bones (6/40).
The treatment protocols were based on previously established protocols for castration-resistant prostate cancer, with the dose per cycle as 6.0-7.4 GBq in most cases and the number of cycles between one and four (Table III). The duration of the treatment varied from 4 to 11 weeks when more than one cycle was used.
Characteristics of the applied 177Lu-PSMA radioligand treatment, outcome, and toxicity.
Unfortunately, none of the 15 publications indicated the presence/absence of serum PSA decline of ≥50% (Table III). Among the 10 unpublished cases, such decline was detected in seven, while the serum PSA decline did not reach 50% in three cases. Among 35 cases with reported toxicity estimation, the reaction of only one patient was estimated as high toxicity, and for the other 34 patients, toxicity was low. The outcome was reported for only 28 patients out of 40. Of them, seven patients died, with a survival range between 2 weeks and 9 months, but no sound estimation is possible because various cancer types were involved. The remaining 12 patients survived to the time of publication or retrieval, with a follow-up period from 15 weeks to 3 years.
Discussion
We aimed to collect all available published, conference-reported, or unpublished cases in which PSMA-targeted radioligand therapy was used to treat non-prostatic cancer. Only a few patients have been treated so far, and these pioneering efforts represent preliminary results. These results indicated that i) yes, 177Lu-PSMA can be used for the treatment of non-prostatic cancer, ii) the toxicity of such treatment is low, and iii) the outcome is relatively good (21 out of 28 patients survived).
Most of the specialists in theranostics who tried 177Lu-PSMA for the treatment of non-prostatic cancer used already well-established and tested protocols and guidelines recommended for treating castration-resistant prostate cancer (49, 50). Therefore, while pioneering, these efforts were relatively safe from the beginning. These protocols and guidelines are constantly improving; the last update was published in 2023 (51). We anticipate more novel applications of 177Lu-PSMA in the future. Specifically for osteosarcoma, no cases were reported in publications, and we obtained data only for unpublished cases. Nevertheless, there are numerous indications in the emerging literature that expression of PSMA in osteosarcoma cases is frequent and well-detectable (52-55). Further efforts to treat osteosarcoma with 177Lu-PSMA may be expected.
The proper selection of patients suitable for treatment will be the main effort if 177Lu-PSMA treatment of non-prostatic cancer intensifies. As we mentioned above, the authors of all cases included in this review detected PSMA immuno-histochemical expression as a part of the selection procedure, but they did not measure serum PSA concentration. This variable is essential, and we believe that PSA concentration expressed (in ng/ml) should be measured as a part of the pretreatment decision-making process. For castration-resistant prostate cancer and especially for metastatic castration-resistant prostate cancer, these concentrations vary widely from 0.1 to 3,121.6 ng/ml (56). For patients with prostate cancer, PSA ≥1.75 ng/ml was suggested to be the optimal value for identifying positive 18F-fluorocholine-PET/CT findings (57). Further research will be needed to establish similar indicators for non-prostatic cancer (16). After that, the selection process of patients with non-prostatic cancer suitable for 177Lu-PSMA treatment will need to be optimized. Further development of this treatment modality for non-prostatic cancer may be expected.
Study limitations. All cases included in this review were small-group or case report studies related to patients with non-prostatic cancer. Further prospective investigations are required to validate our findings in a larger number of patients with various PSMA-expressing cancer types.
In conclusion, PSMA-targeted radioligand therapy was used to treat different non-prostatic cancer types and target other organs in 40 cases. These pioneering efforts indicated that 177Lu-PSMA can be used to treat non-prostatic cancer with PSMA expression. The toxicity of such treatment was low, and the outcome was relatively good.
Data Availability
While a systematic review article, the full dataset is presented in the current submission.
Acknowledgements
The Authors thank Dr. Yael Eshet from Sheba Medical Center, Israel, and Liliia Nehuliaieva, Andrei Krupynskyi Lviv Medical Academy, Lviv, Ukraine, each of whom detected and presented two unpublished cases.
Footnotes
Authors’ Contributions
HG: Conceptualization; methodology; investigation – literature search, screening process, selection, analysis; data curation; writing - review and editing; supervision. KKT: Investigation - literature search, screening process, selection; search for unpublished cases. OV: Investigation – literature search, screening process, selection, analysis; validation; writing – original draft. All Authors approved the final version of the article.
Conflicts of Interest
The Authors have no relevant financial or non-financial interests to disclose.
- Received March 27, 2024.
- Revision received April 21, 2024.
- Accepted April 25, 2024.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).