Introduction and EpidemiologyIn 2018 in the United States, there will be an estimated 164,690 new cases of prostate cancer (19% of all male cancer incident cases, 1st) and an estimated 29,430 prostate cancer mortalities (9% of all male cancer deaths, 2nd only to lung/bronchus cancer).1 For the last 30 or more years, prostate cancer has been the most common noncutaneous malignancy among men in the United States, with 1 in 7 men being diagnosed with the disease.2 De-novo metastatic prostate cancer incidence seems to vary by geographical region and ranges from 4.4 to 9.9 per 100,000 men. A recent study found that over the last several decades, the incidence of de novo metastatic prostate cancer was decreasing in the United States (12.0 to 4.4 per 100,000 men) but increasing in Denmark (6.7 to 9.9 per 100,000 men).3 The exact mechanism for these epidemiologic differences is not clear, but likely related to varying uptake and utilization of PSA surveillance.
With improvements in the treatment of advanced prostate cancer over the last decade, men with advanced disease are living longer and developing non-lymph node visceral metastases. In a single-institution Japanese study (from 2000-2014), among 1,038 prostate cancer patients, there were 144 (19.8 %) men with castration-resistant prostate cancer (CRPC) and 43 (33.1%) patients developing visceral metastases after CRPC progression.4 At diagnosis, the sites of visceral metastases included lung (89.5%), liver (5.3%), and adrenal glands (5.3%). After CRPC progression, new visceral metastases were found in the lung (47.3%), liver (43.6%), and adrenal gland (9.1%). Among 359 CRPC patients in the UK (June 2003 to December 2011), the frequency of radiologically detected visceral metastases before death was 32%; among the 92 patients with a CT scan performed within 3 months of death, 49% had visceral metastases, most commonly involving the liver (20%) and lung (13%).5 These findings confirm a large autopsy study that found among 1,500 prostate cancer patients, 25% of men had liver metastases and 46% had lung metastases.6 Of men participating in first-line studies for metastatic CRPC (mCRPC), ~20% of patients had non-lymph node soft tissue visceral metastases.7,8 As such, leaders in the field have suggested that men with visceral metastases have been an underestimated and understudied subgroup of patients with advanced and heavily treated mCRPC.9,10 The objective of this article is to discuss the biology of visceral metastases, assess the impact of visceral metastases on survival, highlight several large trials that have performed subgroup analyses of visceral metastases patients, and discuss emerging therapeutic regimens for these patients, specifically radioligand targeted therapy.
The Biology of Visceral MetastasesWe are likely only beginning to understand the biology of visceral metastases, particularly as it differs from that of bone metastases. There are several interrelated factors leading to differing pathophysiology between visceral and bone metastases, namely intrinsic cellular factors, the tumor microenvironment, and systemic factors.
1. Cellular Factors: Immunohistochemical analysis of tissue microarrays examining the antiapoptotic pathways expressed in visceral vs bone metastases found that soft-tissue metastases are more likely to express nuclear survivin, whereas bone lesions demonstrate relative overexpression of cytoplasmic survivin, B-cell lymphoma 2, and myeloid cell leukemia 1.11
2. Tumor Microenvironment: microarray studies have found physiologically and clinically important differences between bone, liver, and lymph node metastases. Visceral lesions derived from liver and lymph nodes were found to express an angiogenic profile different from that of liver metastases alone, with significant relative overexpression of the proangiogenic factor angiopoietin-2.12
3. Systemic Factors: serum cytokine levels are associated with prognosis as well as with the presence of liver metastases among prostate cancer patients.9,13 A number of studies have examined levels of TGF-β and interleukin-6 (IL-6) as prognostic markers, finding that the addition of TGF-β and soluble IL-6 receptor levels to a preoperative nomogram significantly improved the ability to predict biochemical progression of the disease.14
Impact of Visceral Metastases on SurvivalPatients with visceral metastases invariably have a worse prognosis than patients with bone-only metastases, likely secondary to an overall increased disease burden.5,10,15-26 In a study including patients in the SEER database (2010-2013), patients with de-novo bone-metastases plus visceral metastases had significantly worse prostate cancer-specific mortality (vs bone only): bone + brain metastases HR 1.48, 95%CI 1.05-2.10; bone + liver metastases HR 2.18, 95%CI 1.79-2.65; bone + lung metastases HR 1.33, 95%CI 1.13-1.56.27
Several large phase III randomized controlled trials (RCTs) have assessed the impact of visceral metastases on survival outcomes using post-hoc analysis of the trial data. The TAX 327 trial found that docetaxel plus prednisone improved OS, pain scores, PSA level, and quality of life compared to mitoxantrone plus prednisone among patients with mCRPC.8 A decade after this publication, Pond et al.25 performed a post-hoc analysis of this data stratified by metastasis site. They found that men with liver metastases with or without other metastases had a shorter median OS (10.0 months; 95%CI 5.4-11.5) than men with lung metastases with or without bone or nodal metastases (median OS: 14.4 months; 95%CI 11.5-22.4). The AFFIRM trial showed that treatment with the androgen receptor inhibitor enzalutamide led to significant improvements in outcomes for patients with mCRPC.28 A subsequent study assessed patients in the AFFIRM trial who had liver and/or lung metastases.20 In patients with liver metastases (n = 92), enzalutamide treatment was associated with a lower risk of radiographic progression (HR 0.645, 95%CI 0.413-1.008), improved 12-month OS (37.7% vs 20.6%) and radiographic progression-free survival (rPFS) (11.6% vs 3.0%) rates compared to those on placebo. Furthermore, patients treated with enzalutamide had higher PSA response rates (35.1% vs 4.8%) compared with placebo. Similarly, patients with lung metastases (n = 104) treated with enzalutamide also had an improved median OS (HR 0.848, 95%CI 0.510-1.410), reduced risk of radiographic progression (HR, 0.386, 95%CI 0.259-0.577), improved 12-month OS (65.1% vs 55.3%) and rPFS (30.9% vs 8.2%) rates, and a better PSA response rate (52.1% vs 4.9%) compared with those who received placebo. The PREVAIL clinical trial tested enzalutamide in men with mCRPC prior to chemotherapy, finding a decreased risk of radiographic progression and death among those taking enzalutamide compared to placebo.29 Of the 1,717 patients in PREVAIL, 12% had visceral metastases: 74 with liver-only or liver/lung metastases and 130 with lung only metastases.19 In patients with liver metastases, treatment with enzalutamide was associated with an improvement in rPFS (HR 0.44, 95%CI 0.22-0.90) but not OS. Among patients with lung metastases only, enzalutamide significantly improved rPFS (HR 0.14, 95%CI 0.06-0.36) and OS (HR 0.59, 95%CI 0.33-1.06). Patients with liver metastases had worse outcomes than those with lung metastases, regardless of treatment.
Results of post-hoc analyses of phase III RCTs showing poor outcomes among patients with visceral metastases have also been confirmed using population-level studies. Gandaglia et al.24 utilized the SEER-Medicare database (1991-2009) to assess outcomes of 3,857 patients presenting with metastatic prostate cancer. Among these patients, 80.2% had bone metastases, 10.9% had bone plus visceral metastases, 6.1% had visceral only metastases, and 2.8% had lymph node only metastases. Patients with bone plus visceral metastases had the worst cancer-specific survival (median 19 months), following by visceral only metastases (median 26 months), bone-only metastases (median 32 months) and lymph node only metastases (median 61 months). Patients with visceral metastases had a significantly higher risk of overall and cancer-specific mortality compared to those with exclusively lymph node metastases (p<0.001), and the unfavorable impact of visceral metastases persisted in the oligometastatic subgroup. Whitney et al.18 studied 494 men with M0 CRPC (diagnosed after 1999) from five Veterans Affairs hospitals in the Shared Equal Access Regional Cancer Hospital (SEARCH) database who later developed metastases. Among these patients, 236 men had a CT scan performed, of which 38 (16%) had visceral metastases, including 19 patients with liver metastases, 8 patients with lung metastases, and 16 patients with other locations of metastases. The authors found that visceral metastases were a predictor of OS on univariate analysis and after risk adjustment (HR 1.84, 95%CI 1.24-2.72).
To further assess the impact of metastatic site on OS among men with mCRPC, a collaborative group performed an individual patient data meta-analysis of 8,820 men with mCRPC who received docetaxel chemotherapy in nine phase III RCTs.22 Site of metastases was categorized as lymph node only, bone with or without lymph node involvement (with no visceral metastases), and lung metastases (but no liver), and any liver metastases. 72.8% of patients had a bone with or without lymph node metastases, 20.8% had a visceral disease, and 6.4% had lymph node-only disease. Men with lymph node-only disease had the best survival with a median OS of 31.6 months, followed by men with non-visceral bone metastases (median OS 21.3 months), men lung metastases (median OS 19.4 months), and those with liver metastases (median OS 13.5 months).
There are several take-home messages from these studies assessing survival outcomes among patients with visceral metastases:
1. Patients with any degree of liver metastases typically have the worst survival outcomes compared to those with bone metastases or other sites of visceral metastases
2. Patients with visceral metastases do have a response to enzalutamide (either in the pre- or post-chemotherapy setting), although their prognosis remains poor
Radioligand Therapy for mCRPC PatientsThe recent uptake in the utilization of PSMA-PET/CT imaging has led to a new field of therapy among heavily pretreated mCRPC patients: radioligand directed therapy. The high PSMA expression in prostate cancer metastases makes it a promising approach to developing new tracers for targeted radionuclide therapies. Since 2015, several institutional studies have reported promising results for response rates and a favorable safety profile after radioligand therapy with 177Lu-PSMA-617 in patients with mCRPC,30-34, however, these studies have suffered from small sample sizes and thus poor generalizability. In an effort to overcome these issues, Rahbar and colleagues35 performed a multicenter German analysis among a cohort of patients treated with 177Lu-PSMA-617. There were 145 patients with mCRPC treated with 177Lu-PSMA-617 at 12 centers undergoing 1-4 therapy cycles with an activity range of 2-8 GBq per cycle. Among these patients, 87% had bone, 77% lymph node, 20% liver, 14% lung, and 2% other sites of metastases. The study reported an overall biochemical response rate of 45% after all therapy cycles, including 40% of patients who responded after a single cycle. Notably, negative predictors of biochemical response include elevated alkaline phosphatase and the presence of visceral metastases.
A study published last month reported on 100 consecutive patients at a single institution receiving 177Lu-PSMA-I&T, treated with a median of two cycles of therapy (range 1-6).36 Among these 100 patients, 57 had received ≥3 prior treatment regimens for mCRPC. There were 87 patients that had lymph metastases and 35 with visceral metastases, including 18 with liver, 11 with lung and 8 with adrenal metastases. A PSA decline of ≥50% was achieved in 38 patients, the median clinical progression-free survival was 4.1 months, and median OS was 12.9 months. The presence of visceral metastasis was the only variable associated with a poor PSA response (p = 0.049), as only nine of 35 (26%) patients with visceral metastasis achieved a maximum PSA decline of ≥50%. The authors concluded that the presence of visceral metastases and rising LDH were associated with worse treatment outcome.
Although 177Lu-PSMA-617 is the most well-studied radioligand to date, there are several other compounds in development and undergoing initial testing. These compounds include: 177Lu-J591, 90Y-J591, 131I-MIP 1095, 177Lu-PSMA-I&T, and 225Ac-PSMA-617.37
ConclusionsSecondary to the improved treatment options available for patients with mCRPC, these men are living longer and thus increasing the prevalence of mCRPC patients with visceral metastases. Although post-hoc studies of enzalutamide trials in the pre- and post-chemotherapy mCRPC setting demonstrate a degree of response, visceral metastases are associated with poor survival outcomes. Initial radioligand therapy studies, primarily with 177Lu-PSMA-617, show promise for heavily treated mCRPC patients, although subgroup analyses of these studies also demonstrate worse survival among patients with visceral metastases. For the future design of phase II and phase III clinical trials among men with mCRPC, patients should be stratified by metastasis site to preclude patients with visceral metastases being inadvertently randomized to an unbalanced trial arm. Further efficacious treatment options for these patients are in dire need. The treatment of visceral metastases is one of the new therapeutic frontiers for prolonging not only quantity but also the quality of life.
Published Date: April 16th, 2019