Visual Abstract
Abstract
177Lu-prostate-specific membrane antigen (PSMA) is an effective treatment for metastatic castration-resistant prostate cancer. Rarer treatment-related adverse events have not yet been described. Methods: We present case reviews of 2 men with a marked hypocalcemic osteosclerotic response to 177Lu-PSMA-I&T therapy. A clinical dataset of 177Lu-PSMA-I&T therapy was evaluated to estimate the incidence and clinical association with hypocalcemia. Results: Forty-one of the 127 men (32%) had a serum calcium drop, and 6 (5%) developed clinical hypocalcemia during 177Lu-PSMA therapy. The baseline total tumor volume was significantly higher in those who developed hypocalcemia (median, 3,249 cm3 [interquartile range, 1,856–3,852] vs. 465 [interquartile range 135–1,172]; P = 0.002). The mean prostate-specific antigen response in those with hypocalcemia was 78% (SD, 24%). Conclusion: Hypocalcemia may occur in response to 177Lu-PSMA-I&T, particularly with both high-volume bone metastases and a significant prostate-specific antigen response, and may be severe, requiring corticosteroids. Further evaluation of 177Lu-PSMA–induced hypocalcemia is required to better understand mechanisms, optimal treatments, and repercussions from any subsequent osteosclerotic response.
In metastatic castration-resistant prostate cancer (mCRPC), 177Lu-prostate-specific membrane antigen (PSMA) is an effective treatment that is generally associated with few adverse events (1,2). We report 2 cases in which men treated with 177Lu-PSMA-I&T for mCRPC developed clinically significant hypocalcemia on therapy; examine the incidence of 177Lu-PSMA-I&T–associated hypocalcemia in a clinical dataset; and discuss potential mechanisms, treatment strategies, and potential long-term consequences of this phenomenon.
MATERIALS AND METHODS
Men with mCRPC who had disease progression despite at least 1 line of androgen receptor signaling inhibition and who were either ineligible for taxane chemotherapy or for whom it had failed were considered for treatment with 177Lu-PSMA-I&T in a clinical treatment program (2022/ETH00924). All men received a minimum of 2 doses of 177Lu-PSMA-I&T at 6-wk intervals. A median of 8.0 GBq (interquartile range [IQR], 8.0–8.5 GBq) was administered at each dose via a slow intravenous injection. Before each dose and at 3-wk intervals, blood was collected for biomarkers including hemoglobin, platelets, lactate dehydrogenase, calcium, alkaline phosphatase, albumin, and prostate-specific antigen (PSA).
RESULTS
Case One
A 71-y-old man with progressive, symptomatic mCRPC and widespread, highly PSMA-avid bony metastases on 68Ga-PSMA-11 PET/CT commenced treatment with 8.5 GBq of 177Lu-PSMA-I&T and showed a marked biochemical response (PSA, 540 to 31 μg/L) after 4 doses 6 wk apart. He received 2 additional doses after a treatment break. He was originally diagnosed 11 mo beforehand with high-volume de novo metastatic prostate adenocarcinoma and progressed on 3 lines of systemic therapy (Tables 1 and 2. He was on goserelin and denosumab, 120 mg monthly, at the initiation of 177Lu-PSMA-I&T. He had normal serum calcium concentrations during the 11 mo that he was taking denosumab before 177Lu-PSMA-I&T. He developed hypocalcemia after the first dose of 177Lu-PSMA-I&T, which worsened despite calcium replacement and cessation of denosumab (Fig. 1). The hypocalcemia was associated with an increased alkaline phosphatase level and serum N-terminal propeptide of type 1 collagen (P1NP) concentration (744 μg/L; reference range, 15–115 μg/L; Fig. 1). The serum parathyroid hormone concentration was appropriately elevated (27.7 pmol/L; reference range, 1.6–7.2). He commenced prednisone, 60 mg daily, to suppress the exaggerated osteoblastic response, in addition to calcitriol and an increased dose of calcium carbonate. The level of serum P1NP normalized within 6 wk of commencing prednisone, whereas the level of serum C-terminal telopeptide of type 1 collagen (CTx) did not increase until 10 mo after 177Lu-PSMA-I&T commencement (to 2,150 ng/L; reference range, 100–750 ng/L), consistent with a postdenosumab cessation rebound (Fig. 1). A baseline dual-energy x-ray absorptiometry bone mineral density scan at the onset of hypocalcemia demonstrated a generalized high bone mass (lumbar spine T-score, +9.5; left femoral neck T-score, +4.2), and serial 177Lu-PSMA SPECT/CT imaging showed persistently high-volume PSMA-avid disease (Fig. 2). A tetracycline-labeled transiliac bone biopsy was attempted; however, the results were nondiagnostic because of difficulty in sampling the sclerotic bone. He remains clinically well and normocalcemic on prednisone, 10 mg; calcium carbonate, 1,200 mg twice per day; and calcitriol, 0.25 μg twice per day, after 6 doses of 177Lu-PSMA-I&T (total dose, 47.5 GBq) and 18 mo after his initial dose. He commenced treatment with 225Ac-PSMA-617 on a trial basis at clinical and radiographic progression, 17 mo after 177Lu-PSMA-I&T initiation.
Case Characteristics at Initiation of 177Lu-PSMA-I&T
Case Characteristics During 177Lu-PSMA-I&T Treatment
Case 1, showing first serum calcium (A) and serum alkaline phosphatase (B) for 6 mo before and during 177Lu-PSMA therapy and second serum P1NP (C) and serum CTx (D) during hypocalcemic response to 177Lu-PSMA and recovery phase.
(A and B) Case 1, treated with 177Lu-PSMA-I&T dose 1 (A) (PSA, 540 μg/L, and total tumor volume, 1,923 cm3, on 177Lu SPECT) and dose 2 (B), with marked PSA response (PSA, 31 μg/L) but persistent high-volume disease on 177Lu-PSMA SPECT (total tumor volume, 1,606 cm3). (C and D) Case 2, treated with 177Lu-PSMA-I&T dose 1 (C) (PSA, 4,960 μg/L, and total tumor volume, 3,453 cm3) and dose 2 (D), with marked PSA response (PSA, 723 μg/L) and significant reduction in total tumor volume (1,247 cm3). Both patients developed clinically significant hypocalcemia after dose 1 177Lu-PSMA, requiring high-dose steroid treatment.
Case Two
An 81-y-old man with progressive mCRPC and intensely PSMA-avid large-volume skeletal metastases on 68Ga-PSMA-11 PET/CT commenced 177Lu-PSMA-I&T after progression despite multiple lines of treatment (Tables 1 and 2). At 177Lu-PSMA-I&T initiation, he was on goserelin but had ceased taking denosumab (120 mg monthly) for transient hypocalcemia that had resolved with calcium carbonate, 600 mg daily, and calcitriol, 0.25 μg twice per day. Despite ongoing cessation of denosumab for more than 6 wk, he presented with fatigue and severe hypocalcemia (1.68 mmol/L) 1 wk after his first dose of 177Lu-PSMA-I&T. The parathyroid hormone level was appropriately elevated (42 pmol/L). Calcium carbonate was increased to 1,200 mg 3 times per day, and the hypocalcemia resolved. Severe hypocalcemia recurred 1 mo later—again, after the second dose of treatment. These recurrent episodes were managed with intravenous calcium gluconate, reuptitration of calcium carbonate, 50 mg of prednisone daily, and cessation of 177Lu-PSMA-I&T. He had a good biochemical response to treatment (PSA, from 4,960 to 730 ng/mL), with a significant reduction of total tumor volume on SPECT (Fig. 2). However, he required a slow prednisone wean with ongoing calcium carbonate, 1,200 mg daily, and calcitriol, 0.25 μg twice per day. He developed PSA progression and clinical progression 4 mo after his initial 177Lu-PSMA-I&T, having received only 2 doses because of recurrent hypocalcemia requiring hospitalization.
Clinical Dataset
Review of a clinical dataset of 127 patients treated with 177Lu-PSMA-I&T (3) demonstrated that 41 (32%) had a fall in serum calcium of any magnitude at any point between starting treatment and administration of dose 3 (12 wk). Six patients (5%) developed new-onset hypocalcemia (serum corrected calcium falling below the reference range) in the first 12 wk of treatment. The mean PSA response in those with hypocalcemia was 78% (SD, 24%), with 1 patient showing no PSA response to treatment. Baseline SPECT total tumor volume was significantly higher in those who had a reduction in calcium than in those who maintained their calcium level (median, 1,017 cm3 [IQR, 331–1,831 cm3] vs. 369 cm3 (IQR, 96–1,035 cm3); P = 0.01) and in those who developed hypocalcemia than in those who did not (median, 3,249 cm3 (IQR, 1,856–3,852 cm3) vs. 465 cm3 (IQR, 135–1,172 cm3); P = 0.002) (Fig. 3). Some patients with hypocalcemia after treatment developed marked osteosclerosis subsequently, despite an excellent biochemical and imaging treatment response (Fig. 4).
Box plot showing relationship between first-dose 177Lu-PSMA-I&T SPECT total tumor volume and development of posttreatment hypocalcemia. Baseline total tumor volume was significantly higher in patients who developed posttreatment hypocalcemia (P = 0.002).
177Lu-PSMA SPECT/CT images for 1 of 6 cases treated with 177Lu-PSMA-I&T who developed hypocalcemia. Images within 24 h after dose 1 (left) and dose 6 (right) show significant reduction in cancer volume with concomitant marked osteosclerotic response (arrows).
DISCUSSION
To our knowledge, this is the first report of clinically significant hypocalcemia developing in response to successful 177Lu-PSMA-I&T treatment. All men impacted had high-volume disease at baseline, and the 2 cases we presented experienced more than a 90% reduction in PSA in response to 177Lu-PSMA-I&T. Tyr is a Norse god who sacrificed his hand to control a powerful wolf god. Similarly, hypocalcemia and osteosclerosis appears to be an adverse consequence of an excellent treatment response.
On the basis of the biochemistry results in our cases, the severe hypocalcemia was independent of parathyroid hormone or systemic controls of serum calcium. The hypocalcemia appeared to be triggered by a direct local tumor effect on bone by 177Lu-PSMA rather than denosumab, given that both patients had normal calcium levels until receiving 177Lu-PSMA-I&T. In case 1, serum alkaline phosphatase and P1NP were markedly elevated at the onset of severe hypocalcemia, indicating an exaggerated osteoblastic response. Serum CTx (a bone resorption marker) was within the reference range but was higher than expected for a patient receiving high-dose denosumab, given the fact that denosumab suppresses CTx within days and remains suppressed during treatment (4,5). Given the markedly exaggerated P1NP levels in relation to CTx, we hypothesize that the 177Lu-PSMA-I&T triggered an uncoupling of bone turnover favoring bone formation over resorption. The resulting increased skeletal demand for calcium for bone formation and mineralization depleted the circulating calcium stores, resulting in severe hypocalcemia similar to that seen in hungry bone syndrome (6).
Prostate cancer is known to cause sclerotic bone metastases, with an increase in sclerotic foci well recognized in response to effective therapy (7). This phenomenon is a major reason that CT is not used to analyze bone lesions on RECIST and that new lesions found on bone scans during the first few months of treatment are not considered to be disease progression. However, the mechanism by which prostate cancer induces osteosclerosis in response to treatment is not fully understood, and although significant concomitant hypocalcemia has been reported with rapid cancer progression (8), it has not been described after exceptional responses to treatment. Treatment-induced hypocalcemia in prostate cancer has been reported as an adverse reaction to only high-dose denosumab and bisphosphonates (5).
With increasing use of 177Lu-PSMA therapy, this is an important phenomenon to recognize, particularly as it may require escalated therapy to manage, such as glucocorticoid therapy. In the patients who had an unrecognized hypocalcemic response to 177Lu-PSMA-I&T in our clinical dataset, there was a significant increase in sclerotic disease volume. This increase has the potential to reduce the available marrow reserve and may lead to early induction of marrow failure with subsequent treatments, particularly with 177Lu-PSMA therapy potentially moving to earlier in the disease course.
Supraphysiologic glucocorticoid administration suppresses bone formation by inhibiting osteoblast differentiation and proliferation and inducing osteoblast apoptosis (9). Hence, management of these patients required high-dose glucocorticoids to dampen the exaggerated osteoblastic response and help normalize calcium levels. The underlying mechanism by which 177Lu-PSMA drives this exaggerated osteoblastic response and results in hypocalcemia is unclear and may be the release of osteoblastogenic growth factors by metastatic prostate cancer cells and cells within the bone tumor microenvironment. Further prospective evaluation of such cases, with close monitoring of calcium and bone turnover markers and tetracycline-labeled bone biopsy using techniques such as micro-CT, histomorphometry, and immunohistochemistry, will likely enhance understanding and identify factors that select patients who need prophylactic treatment, thereby preventing the hypocalcemic and sclerotic events.
CONCLUSION
Clinically significant hypocalcemia and osteosclerosis are a rare but important side effect of 177Lu-PSMA-I&T therapy in men with high-volume osseous metastatic disease and a significant treatment response. Importantly, calcium and calcitriol supplementation were insufficient in managing the hypocalcemia, and oral prednisone was required. Further work delineating the mechanism in the tumor microenvironment is required to fully understand and develop effective management options for patients with this phenomenon.
DISCLOSURE
No potential conflict of interest relevant to this article was reported.
KEY POINTS
QUESTION: Are good responders to 177Lu-PSMA at risk for clinically significant hypocalcemia?
PERTINENT FINDINGS: We found that patients with high-volume bone metastases and a significant PSA response are at a higher risk of developing hypocalcemia.
IMPLICATIONS FOR PATIENT CARE: Although hypocalcemia is a rare side effect of 177Lu-PSMA, it can be clinically significant and may require treatment.
ACKNOWLEDGMENTS
We thank the patients, as well as the clinical staff at the Department of Theranostics and Nuclear Medicine of St Vincent’s Hospital, Sydney, Australia, for their support.
Footnotes
Published online Jun. 22, 2023.
- © 2023 by the Society of Nuclear Medicine and Molecular Imaging.
REFERENCES
- Received for publication March 23, 2023.
- Revision received May 10, 2023.
