Abstract
To report the regional locations of metastases and to estimate the prognostic value of the pattern of regional metastases in men with metastatic hormone-sensitive prostate cancer (mHSPC), we retrospectively analyzed 870 mHSPC patients between November 28, 2009, and February 4, 2021, from West China Hospital in Chengdu, China. The patients were initially classified into 5 subgroups according to metastatic patterns as follows: simple bone metastases (G1), concomitant bone and regional lymph node (LN) metastases (G2), concomitant bone and nonregional LN (NRLN) metastases (G3), lung metastases (G4), and liver metastases (G5). In addition, patients in the G3 group were subclassified as G3a and G3b based on the LN metastatic plane (below or above the diaphragm, respectively). The associations of different metastatic patterns with castration-resistant prostate cancer-free survival (CFS) and overall survival (OS) were analyzed by univariate and multivariate analyses. The results showed that patients in G1 and G2 had relatively favorable clinical outcomes, patients in G3a and G4 had intermediate prognoses, and patients in G3b and G5 had the worst survival outcomes. We observed that patients in G3b had outcomes comparable to those in G5 but had a significantly worse prognosis than patients in G3a (median CFS: 8.2 months vs 14.3 months, P = 0.015; median OS: 38.1 months vs 45.8 months, P = 0.038). In conclusion, metastatic site can predict the prognosis of patients with mHSPC, and the presence of concomitant bone and NRLN metastases is a valuable prognostic factor. Furthermore, our findings indicate that the farther the NRLNs are located, the more aggressive the disease is.
Keywords: lymph node metastases, metastatic hormone-sensitive prostate cancer, metastatic pattern, metastatic site, nonregional lymph node metastases, staging
INTRODUCTION
Prostate cancer (PCa) is the most commonly diagnosed cancer in male patients.1 Approximately 6% of PCa patients in the USA and 25%–44% of PCa patients in Asia present with metastatic diseases at initial diagnosis.1,2 Metastatic prostate cancer represents a highly heterogeneous entity characterized by variable survival outcomes, and the metastatic pattern is one of the influencing factors.3,4 PCa has different patterns of metastatic dissemination, including bone, lymph node (LN), lung, and liver metastases, which occur alone or concomitantly. Based on the presence and number of bone and visceral metastases, previous studies defined “low tumor burden” and “high tumor burden” disease to guide clinical treatment.5,6 However, these definitions do not consider LN metastases. Although LN metastases have been reported as a low-risk factor compared to bone or visceral metastases,3,4 the presence of concomitant bone and LN metastases is not uncommon.3,7 Several studies have indicated that PCa patients with concomitant bone and nonregional lymph node (NRLN) metastases have poorer survival outcomes than those with only bone metastases.8,9 However, the prognostic importance of the regional location of LN metastases has not been fully evaluated.
Therefore, this study aimed to report the regional locations of metastases and to estimate the prognostic value of the regional location of LN metastases in men with metastatic hormone-sensitive prostate cancer (mHSPC).
PATIENTS AND METHODS
Study design
Between November 25, 2009, and February 4, 2021, a total of 870 de novo mHSPC patients from the West China Hospital (Chengdu, China) prostate cancer database were included in this study. All patients underwent pelvic magnetic resonance imaging, bone scans, and enhanced computed tomography (CT) of the chest and abdomen at the initial diagnosis. The quality of the CT images was evaluated by two experienced radiologists (Yun-Tian Chen and Jing Yao, West China Hospital, Sichuan University) according to a CT quality scoring system (Supplementary Table 1), and all the CT images were assessed as having moderate to high quality (score ≥3). The metastatic pattern was used to classify patients as follows: simple bone metastases (G1), concomitant bone and regional LN (RLN) metastases (G2), concomitant bone and nonregional LN (NRLN) metastases (G3), lung metastases (G4), and liver metastases (G5). In addition, patients in the G3 group were further subclassified as G3a and G3b based on the regional location of the LN metastases (below or above the diaphragm). Details of the classification scheme are provided in Supplementary Table 2. Notably, only 4 and 3 patients were diagnosed with brain and only NRLN metastases, respectively, in our database, so they were excluded.
Supplementary Table 1.
Computed tomography image quality scoring system
| Score | Scoring rules |
|---|---|
| 1 | No diagnostic value at all |
| 2 | Poor SNR, strong artifacts and poor diagnostic value |
| 3 | Low SNR, some artifacts and declined diagnostic value |
| 4 | Slightly low SNR, few artifacts and relatively good diagnostic value |
| 5 | Good SNR, almost no artifacts and very good diagnostic value |
SNR: signal noise ratio
Supplementary Table 2.
Classification of site of metastases
| Category | Site of metastases | Presence of RLN metastases | Presence of NRLN metastases | Presence of bone metastases | Presence of lung metastases | Presence of liver metastases |
|---|---|---|---|---|---|---|
| G1 | Bone only | No | No | No | No | No |
| G2 | Bone with RLN | Yes | No | No | No | No |
| G3 | Bone with NRLN | Yes/No | Yes | Yes | No | No |
| G4 | Lung | Yes/No | Yes/No | Yes/No | Yes | No |
| G5 | Liver | Yes/No | Yes/No | Yes/No | Yes/No | Yes |
| Category | Classification of patients with G3, site of LN | |||||
| G3a | Subdiaphragmatic LN metastases, including LN in the area of hepatic vessel, abdominal vessel or retroperitoneum | |||||
| G3b | Supradiaphragmatic LN metastases, including LN in the area of mediastinum, hilar, supraclavicular, mandible or neck | |||||
LN: lymph node; RLN: regional LN; NRLN: nonregional lymph nodes
The Institutional Review Board (IRB) in West China Hospital approved this study (approval No. 20211703). Consent was obtained from all patients. We collected the following characteristics at initial diagnosis: age, Eastern Cooperative Oncology Group (ECOG) score, Gleason score (GS), prostate-specific antigen (PSA) level, lactate dehydrogenase (LDH) level, hemoglobin (HGB) level, alkaline phosphatase (ALP) level, and treatments during mHSPC. The definitions of “high burden” and “high risk” were based on the CHAARTED and LATITUDE studies, respectively.10,11
Endpoints
Castration-resistant prostate cancer-free survival (CFS) and overall survival (OS) were the study endpoints, and the last follow-up was on September 15, 2021. CFS was defined as the period from the initial diagnosis to the occurrence of castration-resistant prostate cancer (CRPC). CRPC was defined by the 2020 European Association of Urology (EAU) guidelines.12,13 The definition of OS was the time from the initial diagnosis to death.
Statistical analyses
We applied the Chi-square test to compare the baseline characteristics. The Kaplan–Meier method and log-rank test were used to compare CFS and OS. Univariate Cox analysis was applied to evaluate the value of each clinicopathological factor in predicting CFS and OS. Factors with P < 0.05 were further analyzed in multivariate Cox analysis. The prognostic prediction accuracy was evaluated by the concordance index (C-index).
Statistical analyses were performed with SPSS 25.0 (IBM SPSS Inc., Chicago, IL, USA) and R software 4.1.0 (R Core Team 2021, Vienna, Austria). Statistical significance was defined as P < 0.05.
RESULTS
Patient characteristics
The baseline characteristics of 870 mHSPC patients are summarized in Table 1. There were 605 (69.5%) patients diagnosed with only bone metastases (G1), 92 (10.6%) patients diagnosed with concomitant bone and RLN metastases (G2), 95 (10.9%) patients diagnosed with concomitant bone and NRLN metastases (G3), 44 (5.1%) patients diagnosed with lung metastases (G4), and 34 (3.9%) patients diagnosed with liver metastases (G5). Among the patients in G3, 75 (8.6%) men were subclassified into G3a, and 20 (2.3%) men were subclassified into G3b. Patients in G1 were associated with lower GS. A total of 775 patients (89.1%) received maximal androgen blockade (MAB; luteinizing hormone-releasing hormone agonist or orchiectomy plus bicalutamide), and 95 patients (10.9%) received androgen deprivation therapy (ADT) plus abiraterone (ABI) or ADT plus docetaxel (DOC) as the first-line treatment. A higher proportion of patients in G2 (18.5%) or G3 (16.8%) received ABI/DOC. The other baseline factors were balanced among the groups (Table 1). After a median follow-up period of 42.1 months, 644 (74.0%) patients progressed to CRPC, and 299 (34.4%) patients died. Overall, the median CFS (mCFS) and median OS (mOS) were 17.6 months and 64.1 months, respectively.
Table 1.
Baseline characteristics of the total cohort
| Variable | Total (n=870) | G1 (n=605) | G2 (n=92) | G3 (n=95) | G4 (n=44) | G5 (n=34) | P |
|---|---|---|---|---|---|---|---|
| Age (year), median (IQR) | 71.0 (65.0–77.0) | 72.0 (66.0–77.0) | 71.0 (65.8–76.3) | 69.0 (61.5–75.0) | 71.0 (65.8–76.0) | 69.0 (63.0–74.5) | 0.168 |
| <71, n (%) | 391 (44.9) | 256 (42.3) | 45 (48.9) | 52 (54.7) | 21 (47.7) | 17 (50.0) | |
| ≥71, n (%) | 479 (55.1) | 349 (57.7) | 47 (51.1) | 43 (45.3) | 23 (52.3) | 17 (50.0) | |
| GS, n (%) | 0.007 | ||||||
| 6–7 | 107 (12.3) | 84 (13.9) | 8 (8.7) | 7 (7.4) | 7 (15.9) | 1 (2.9) | |
| 8 | 198 (22.8) | 150 (24.8) | 12 (13.0) | 21 (22.1) | 8 (18.2) | 7 (20.6) | |
| 9–10 | 565 (64.9) | 371 (61.3) | 72 (78.3) | 67 (70.5) | 29 (65.9) | 26 (76.5) | |
| ECOG score, n (%) | 0.268 | ||||||
| 0–1 | 746 (85.7) | 524 (86.6) | 82 (89.1) | 77 (81.1) | 35 (79.5) | 29 (85.3) | |
| ≥2 | 124 (14.3) | 81 (13.4) | 10 (10.9) | 18 (18.9) | 9 (20.5) | 5 (14.7) | |
| PSA (ng ml−1), mean (range) | 199.5 (4.1–4052.0) | 193.1 (4.1–4052.0) | 201.9 (6.0–2432.0) | 225.7 (4.5–3008.0) | 210.2 (4.2–2553.0) | 218.1 (4.6–2462.0) | 0.098 |
| <100, n (%) | 318 (36.6) | 229 (37.9) | 24 (26.1) | 31 (32.6) | 21 (47.7) | 13 (38.2) | |
| ≥100, n (%) | 552 (63.4) | 376 (62.1) | 68 (73.9) | 64 (67.4) | 23 (52.3) | 21 (61.8) | |
| HGB (g l−1), median (IQR) | 131.0 (120.0–142.0) | 131.0 (121.0–141.0) | 130.8 (119.0–146.0) | 131.0 (120.0–142.0) | 135.5 (125.5–142.0) | 124.0 (110.0–141.5) | 0.293 |
| <120, n (%) | 200 (23.0) | 132 (21.8) | 25 (27.2) | 23 (24.2) | 8 (18.2) | 12 (35.3) | |
| ≥120, n (%) | 670 (77.0) | 473 (78.2) | 67 (72.8) | 72 (75.8) | 36 (81.8) | 22 (64.7) | |
| LDH (IU l-1), median (IQR) | 198.0 (172.0–224.0) | 198.0 (173.0–225.0) | 192.8 (169.5–225.0) | 200.0 (174.5–220.5) | 192.5 (165.1–206.6) | 200.0 (169.5–237.3) | 0.258 |
| <250, n (%) | 736 (84.6) | 509 (84.1) | 81 (88.0) | 79 (83.2) | 41 (93.2) | 26 (76.5) | |
| ≥250, n (%) | 134 (15.4) | 96 (15.9) | 11 (12.0) | 16 (16.8) | 3 (6.8) | 8 (23.5) | |
| ALP (IU l−1), median (IQR) | 101.0 (76.0–159.8) | 104.0 (77.0–164.0) | 96.0 (77.9–134.0) | 94.0 (69.5–146.0) | 102.0 (77.8–142.0) | 106.8 (70.8–244.6) | 0.145 |
| <160, n (%) | 652 (74.9) | 445 (73.6) | 77 (83.7) | 74 (77.9) | 34 (77.3) | 22 (64.7) | |
| ≥160, n (%) | 218 (25.1) | 160 (26.4) | 15 (16.3) | 21 (22.1) | 10 (22.7) | 12 (35.3) | |
| Treatments, n (%) | 0.015 | ||||||
| MAB | 775 (89.1) | 549 (90.7) | 75 (81.5) | 79 (83.2) | 42 (95.5) | 30 (88.2) | |
| ABI/DOC | 95 (10.9) | 56 (9.3) | 17 (18.5) | 16 (16.8) | 2 (4.5) | 4 (11.8) | |
| LATITUDE, n (%) | 0.001 | ||||||
| Low risk | 250 (28.7) | 183 (30.2) | 28 (30.4) | 32 (33.7) | 6 (13.6) | 1 (2.9) | |
| High risk | 620 (71.3) | 422 (69.8) | 64 (69.6) | 63 (66.3) | 38 (86.4) | 33 (97.1) | |
| CHARRTED, n (%) | <0.001 | ||||||
| Low burden | 406 (46.7) | 311 (51.4) | 48 (52.2) | 47 (49.5) | 0 (0) | 0 (0) | |
| High burden | 464 (53.3) | 294 (48.6) | 44 (47.8) | 48 (50.5) | 44 (100.0) | 34 (100.0) |
G1: simple bone metastases; G2: concomitant bone and RLN metastases; G3: concomitant bone and NRLN metastases; G4: lung metastases; G5: liver metastases. LN: lymph node; RLN: regional LN; NRLN: non-RLN; IQR: interquartile range; GS: Gleason score; ECOG: Eastern Cooperative Oncology Group; PSA: prostate-specific antigen; HGB: hemoglobin; LDH: lactate dehydrogenase; ALP: alkaline phosphatase; ABI: abiraterone; DOC: docetaxel; MAB: maximum androgen blockade
Prognostic value of metastatic site in mHSPC patients
As shown in Figure 1, patients in G1 (mCFS: 18.7 months, mOS: 70.5 months) and G2 (mCFS: 23.0 months, mOS: 65.5 months) had relatively favorable survival outcomes among all groups. Patients in G3 (mCFS: 12.3 months, mOS: 43.2 months) and G4 (mCFS: 15.0 months, mOS: 47.7 months) were associated with an intermediate prognosis. Patients in G5 had the worst survival outcomes (mCFS: 8.7 months, mOS: 26.0 months). In addition to metastatic site, univariate Cox regression indicated that ECOG score, GS, HGB, LDH, ALP, and treatments were related to patient prognosis (all P < 0.05; Table 2). After adjusting for clinicopathological characteristics, multivariate Cox analysis further demonstrated that bone with NRLN metastases (CFS: HR [95% CI] = 1.80 [1.41–2.29], P < 0.001; OS: HR [95% CI] = 1.89 [1.33–2.68], P < 0.001), lung metastases (CFS: HR [95% CI] = 1.53 [1.07–2.18], P = 0.019; OS: HR [95% CI] = 1.63 [1.03–2.58], P = 0.036), and liver metastases (CFS: HR [95% CI] = 2.70 [1.89–3.88], P < 0.001; OS: HR [95% CI] = 3.27 [2.15–4.96], P < 0.001) were independent predictors of worse survival outcomes (Table 2).
Figure 1.
Impact of the metastatic site on (a) CFS and (b) OS in patients with mHSPC. G1: simple bone metastases; G2: concomitant bone and RLN metastases; G3: concomitant bone and NRLN metastases; G4: lung metastases; G5: liver metastases. CFS: castration-resistant prostate cancer-free survival; OS: overall survival; mCFS: median CFS; mOS: median OS; mHSPC: metastatic hormone-sensitive prostate cancer; RLN: regional lymph nodes; NRLN: nonregional lymph nodes.
Table 2.
Univariate and multivariate analyses of castration-resistant prostate cancer-free survival and overall survival for all metastatic prostate cancer patients
| Variable | CFS univariate analysis | CFS multivariate analysis | OS univariate analysis | OS multivariate analysis | ||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|||||
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | |
| Age, ≥71 years vs <71 years | 0.88 (0.75–1.02) | 0.094 | 0.89 (0.71–1.12) | 0.331 | ||||
| ECOG, ≥2 months vs <0–1 month | 1.50 (1.21–1.84) | <0.001 | 1.12 (0.90–1.39) | 0.310 | 1.91 (1.46–2.50) | <0.001 | 1.34 (1.01–1.77) | 0.042 |
| GS, 9–10 vs 6–7 | 1.96 (1.51–2.53) | <0.001 | 2.01 (1.55–2.60) | <0.001 | 1.93 (1.35–2.79) | <0.001 | 1.83 (1.26–2.65) | 0.001 |
| GS, 9–10 vs 8 | 1.55 (1.28–1.88) | <0.001 | 1.54 (1.26–1.87) | <0.001 | 2.18 (1.57–3.01) | <0.001 | 2.08 (1.50–2.89) | <0.001 |
| PSA, ≥100 ng ml−1 vs <100 ng ml−1 | 1.13 (0.97–1.33) | 0.126 | 0.97 (0.77–1.23) | 0.787 | ||||
| HGB, ≥120 g l−1 vs <120 g l−1 | 0.59 (0.49–0.70) | <0.001 | 0.69 (0.57–0.83) | <0.001 | 0.53 (0.41–0.68) | <0.001 | 0.64 (0.49–0.83) | 0.001 |
| LDH, ≥250 IU l−1 vs <250 IU l−1 | 1.95 (1.60–2.38) | <0.001 | 1.55 (1.25–1.92) | <0.001 | 2.24 (1.71–2.92) | <0.001 | 1.74 (1.31–2.31) | <0.001 |
| ALP, ≥160 IU l−1 vs <160 IU l−1 | 1.97 (1.66–2.34) | <0.001 | 1.57 (1.30–1.91) | <0.001 | 2.06 (1.62–2.61) | <0.001 | 1.52 (1.17–1.97) | 0.002 |
| ABI/DOC vs MAB | 0.62 (0.46–0.83) | 0.002 | 0.55 (0.41–0.75) | <0.001 | 0.42 (0.21–0.81) | 0.010 | 0.40 (0.20–0.78) | 0.007 |
| G2 vs G1 | 0.90 (0.69–1.18) | 0.434 | 0.97 (0.64–1.47) | 0.874 | ||||
| G3 vs G1 | 1.79 (1.41–2.27) | <0.001 | 1.80 (1.41–2.29) | <0.001 | 1.85 (1.31–2.60) | <0.001 | 1.89 (1.33–2.68) | <0.001 |
| G4 vs G1 | 1.55 (1.09–2.20) | 0.015 | 1.53 (1.07–2.18) | 0.019 | 1.70 (1.08–2.67) | 0.021 | 1.63 (1.03–2.58) | 0.036 |
| G5 vs G1 | 2.70 (1.89–3.86) | <0.001 | 2.70 (1.89–3.88) | <0.001 | 3.50 (2.32–5.29) | <0.001 | 3.27 (2.15–4.96) | <0.001 |
G1: simple bone metastases; G2: concomitant bone and RLN metastases; G3: concomitant bone and NRLN metastases; G4: lung metastases; G5: liver metastases. CRPC: castration-resistant prostate cancer; CFS: CRPC-free survival; OS: overall survival; ECOG: Eastern Cooperative Oncology Group; GS: Gleason score; PSA: prostate-specific antigen; HGB: hemoglobin; LDH: lactate dehydrogenase; ALP: alkaline phosphatase; LN: lymph node; RLN: regional LN; NRLN: non-RLN; ABI: abiraterone; DOC: docetaxel; MAB: maximum androgen blockade; HR: hazard ratio; CI: confidence interval
Prognostic value of the LN metastatic patterns in men with concomitant bone and NRLN metastases
Considering that the sites of NRLN could vary, patients in G3 were subclassified into G3a and G3b based on the LN metastatic pattern (below or above the diaphragm, respectively). Therefore, the prognostic value of LN metastatic patterns was further explored in patients in G3. The baseline clinicopathological features were well balanced between these two groups (Supplementary Table 3). Patients in G3b had significantly worse survival outcomes than those in G3a (mCFS: 8.2 months vs 14.3 months, P = 0.015; mOS: 38.1 months vs 45.8 months, P = 0.038; Figure 2). Multivariate Cox analysis after adjusting for other clinicopathological factors strengthened the above results (Supplementary Table 4). Notably, when we stratified these patients according to the CHAARTED and LATITUDE criteria, no difference in survival outcomes was observed between high- and low-burden patients or between high- and low-risk patients (both P > 0.05; Supplementary Figure 1 (702.8KB, tif) ). The above results indicated that the LN metastatic pattern was an independent prognosticator for patients with concomitant bone and NRLN metastases, and the CHAARTED and LATITUDE criteria did not add a significant prognostic impact.
Supplementary Table 3.
Baseline of the patients with concomitant bone and nonregional lymph node metastases
| Variables | Total (n=95) | G3a (n=75) | G3b (n=20) | P |
|---|---|---|---|---|
| Age (year), median (IQR) | 69.0 (61.5–75.0) | 69.0 (61.5–75.0) | 71.0 (61.5–77.3) | 0.632 |
| <71, n (%) | 52 (54.7) | 42 (56.0) | 10 (50.0) | |
| ≥71, n (%) | 43 (45.3) | 33 (44.0) | 10 (50.0) | |
| GS | 0.865 | |||
| <8, n (%) | 7 (7.4) | 5 (6.7) | 2 (10.0) | |
| =8, n (%) | 21 (22.1) | 17 (22.7) | 4 (20.0) | |
| 9–10, n (%) | 67 (70.5) | 53 (70.7) | 14 (70.0) | |
| ECOG score | 1.000 | |||
| 0–1, n (%) | 76 (80.0) | 60 (80.0) | 16 (80.0) | |
| ≥2, n (%) | 19 (20.0) | 15 (20.0) | 4 (20.0) | |
| PSA (ng ml−1), mean (range) | 225.7 (4.5–3008.0) | 210.1 (4.5–3008.0) | 284.5 (10.4–1352.1) | 0.778 |
| <100, n (%) | 31 (32.6) | 25 (33.3) | 6 (30.0) | |
| ≥100, n (%) | 64 (67.4) | 50 (66.7) | 14 (70.0) | |
| HGB (g l−1), median (IQR) | 131.5 (120.5–143.0) | 131.5 (120.5–143.0) | 125.0 (109.8–137.3) | 0.496 |
| <120, n (%) | 23 (24.2) | 17 (22.7) | 6 (30.0) | |
| ≥120, n (%) | 72 (75.8) | 58 (77.3) | 14 (70.0) | |
| LDH (IU l−1), median (IQR) | 201.0 (176.5–221.5) | 201.0 (176.5–221.5) | 185.5 (164.3–202.9) | 0.671 |
| <250, n (%) | 79 (83.2) | 63 (84.0) | 16 (80.0) | |
| ≥250, n (%) | 16 (16.8) | 12 (16.0) | 4 (20.0) | |
| ALP (IU L−1), median (IQR) | 90.0 (68.5–136.0) | 90.0 (68.5–136.0) | 117.5 (75.3–169.5) | 0.725 |
| <160, n (%) | 74 (77.9) | 59 (78.7) | 15 (75.0) | |
| ≥160, n (%) | 21 (22.1) | 16 (21.3) | 5 (25.0) | |
| Treatments | 0.357 | |||
| MAB, n (%) | 79 (83.2) | 61 (81.3) | 18 (90.0) | |
| ABI/DOC, n (%) | 16 (16.8) | 14 (18.7) | 2 (10.0) | |
| LATITUDE | 0.889 | |||
| High risk, n (%) | 63 (66.3) | 50 (66.7) | 13 (65.0) | |
| Low risk, n (%) | 32 (33.7) | 25 (33.3) | 7 (35.0) | |
| CHARRTED | 0.652 | |||
| High tumor burden, n (%) | 48 (50.5) | 37 (49.3) | 11 (55.0) | |
| Low tumor burden, n (%) | 47 (49.5) | 38 (50.7) | 9 (45.0) |
G3a: concomitant bone and NRLN metastases (below diaphragm); G3b: concomitant bone and NRLN metastases (above diaphragm). LN: lymph node; RLN: regional LN; NRLN: non-RLN; GS: Gleason score; ECOG: Eastern Cooperative Oncology Group; PSA: prostate specific antigen; HGB: hemoglobin; LDH: lactate dehydrogenase; ALP: alkaline phosphatase; ABI: abiraterone; DOC: docetaxel; MAB: maximum androgen blockade; IQR: interquartile range
Figure 2.
Impact of LN metastatic patterns on (a) CFS and (b) OS in patients with concomitant bone and NRLN metastases. G3a: concomitant bone and subdiaphragmatic NRLN metastases; G3b: concomitant bone and supradiaphragmatic NRLN metastases; LN: lymph node; NRLN: nonregional lymph node; CFS: castration-resistant prostate cancer-free survival; OS: overall survival; mCFS: median CFS; mOS: median OS.
Supplementary Table 4.
Univariate and multivariate analyses of castration-resistant prostate cancer free survival and overall survival for patients with concomitant bone and nonregional lymph node metastases
| Variables | CFS univariate analysis | CFS multivariate analysis | OS univariate analysis | OS multivariate analysis | ||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|
|
|||||
| HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | HR (95% CI) | P | |
| Age (year), ≥71 vs <71 | 1.02 (0.65–1.59) | 0.930 | - | - | 0.96 (0.51–1.81) | 0.891 | - | - |
| ECOG (mo), ≥2 vs <0–1 | 1.45 (0.85–2.47) | 0.173 | - | - | 1.76 (0.90–3.45) | 0.097 | - | - |
| GS, 9–10 vs 6–7 | 1.96 (0.71–5.38) | 0.195 | - | - | 1.28 (0.30–5.35) | 0.739 | - | - |
| GS, 9–10 vs 8 | 1.13 (0.66–1.93) | 0.665 | - | - | 1.70 (0.74–3.88) | 0.208 | - | - |
| PSA (ng ml−1), ≥100 vs <100 | 0.87 (0.55–1.39) | 0.569 | - | - | 0.94 (0.50–1.78) | 0.848 | - | - |
| HGB (g l−1), ≥120 vs <120 | 0.83 (0.49–1.40) | 0.482 | - | - | 0.59 (0.27–1.26) | 0.172 | - | - |
| LDH (IU l−1), ≥250 vs <250 | 1.39 (0.78–2.47) | 0.259 | - | - | 0.97 (0.40–2.33) | 0.943 | - | - |
| ALP (IU l−1), ≥160 vs <160 | 1.82 (1.09–3.03) | 0.021 | 2.07 (1.23–3.47) | 0.006 | 1.61 (0.76–3.42) | 0.214 | - | - |
| G3b vs G3a | 1.89 (1.12–3.18) | 0.017 | 1.95 (1.14–3.31) | 0.014 | 2.08 (1.03–4.22) | 0.042 | 2.10 (1.03–4.25) | 0.040 |
| ABI/DOC vs MAB | 0.43 (0.22–0.83) | 0.012 | 0.42 (0.22–0.83) | 0.012 | 0.137 (0.02–1.00) | 0.049 | 0.14 (0.02–0.99) | 0.049 |
G3a: concomitant bone and NRLN metastases (below diaphragm); G3b: concomitant bone and NRLN metastases (above diaphragm). LN: lymph node; RLN: regional LN; NRLN: Non-RLN; CRPC: castration-resistant prostate cancer; CFS: CRPC free survival; OS: overall survival; ABI: abiraterone; DOC: docetaxel; MAB: maximum androgen blockade; HR: hazard ratio; CI: confidence interval; ECOG: Eastern Cooperative Oncology Group; GS: Gleason score; PSA: prostate-specific antigen; HGB: hemoglobin; LDH: lactate dehydrogenase; ALP: alkaline phosphatase
Construction of a predictive model based on metastatic site
As shown in Figure 3a and 3b, due to the overlap in OS for the initial groupings in our study, we grouped G1 and G2 as low risk, G3a and G4 as intermediate risk, and G3b and G5 as high risk. There were significant differences in survival outcomes among these three groups (mCFS: 19.5 months vs 14.9 months vs 8.4 months; mOS: 70.5 months vs 45.8 months vs 28.9 months, all P ≤ 0.001; Figure 3c and 3d). Multivariate Cox regression after adjusting for other clinicopathological characteristics also strengthened this finding (all P ≤ 0.002; Table 3). The new merged model showed stronger predictive power for the prognosis of patients with mHSPC, with the C-index increasing from 0.541 to 0.549 for predicting CFS and from 0.566 to 0.574 for predicting OS.
Figure 3.
Predictive model based on metastatic site. (a) Survival plot of CFS based on G1–G5 model; (b) survival plot of OS based on G1–G5 model; (c) survival plot of CFS based on the new merged model; (d) survival plot of OS based on the new merged model. G1: simple bone metastases; G2: concomitant bone and RLN metastases; G3a: concomitant bone and subdiaphragmatic NRLN metastases; G3b: concomitant bone and supradiaphragmatic NRLN metastases; G4: lung metastases; G5: liver metastases; LN: lymph node; CFS: castration-resistant prostate cancer-free survival; OS: overall survival; mCFS: median CFS; mOS: median OS; RLN: regional lymph nodes; NRLN: non-RLN.
Table 3.
Multivariate analyses of castration-resistant prostate cancer-free survival and overall survival for all metastatic prostate cancer patients (based on merged model)
| Variable | CFS multivariate analysisa | OS multivariate analysisa | ||
|---|---|---|---|---|
|
|
|
|||
| HR (95% CI) | P | HR (95% CI) | P | |
| High risk vs low risk | 2.94 (2.19–3.94) | <0.001 | 3.27 (2.28–4.67) | <0.001 |
| Intermediate risk vs low risk | 1.61 (1.29–2.00) | <0.001 | 1.65 (1.21–2.26) | 0.002 |
| High risk vs intermediate risk | 1.83 (1.30–2.57) | <0.001 | 1.98 (1.28–3.05) | 0.002 |
aAdjusted for ECOG, GS, HGB, LDH, ALP and treatments. Low risk: G1 and G2; intermediate risk: G3a and G4; high risk: G3b and G5; G1: simple bone metastases; G2: concomitant bone and RLN metastases; G3a: concomitant bone and subdiaphragmatic NRLN metastases; G3b: concomitant bone and supradiaphragmatic NRLN metastases; G4: lung metastases; G5: liver metastases. CRPC: castration-resistant prostate cancer; CFS: CRPC-free survival; OS: overall survival; ECOG: Eastern Cooperative Oncology Group; GS: Gleason score; HGB: hemoglobin; LDH: lactate dehydrogenase; ALP: alkaline phosphatase; LN: lymph node; RLN: regional LN; NRLN: non-RLN; HR: hazard ratio; CI: confidence interval
DISCUSSION
In this research, we evaluated the prognostic importance of metastatic patterns in men with mHSPC and explored the prognostic value of LN metastatic patterns in men with concomitant bone and LN metastases.
Our findings indicated that different patterns of metastatic dissemination had a differential prognostic impact on men with mHSPC. Patients with simple bone metastases (G1) and concomitant bone and RLN metastases (G2) were associated with a favorable prognosis, while those harboring concomitant bone and NRLN (G3) and lung metastases (G4) showed intermediate survival outcomes, and patients with liver metastases (G5) had the worst prognosis. Several previous studies reported similar results. Using the SEER database, Ali et al.8 found that patients with concomitant bone and NRLN metastases had a higher risk of death than those with only bone metastases. Another population-based study from The Netherlands also showed that the prognosis of men with concomitant bone and NRLN metastases was between that of men with simple bone metastases and that of men with visceral metastases.9
Furthermore, considering that NRLN could be located at various sites of the body, we first divided the patients in G3 into two subgroups using the diaphragm plane as the boundary. Interestingly, we found heterogeneity between the two groups and that the farther the NRLN was located, the more aggressive the disease was. In fact, similar phenomena have been observed in other types of tumors. Previous studies found that not the number but the location of LNs had an impact on prognosis in thyroid cancer and colorectal cancer.14,15 Anderegg et al.16 demonstrated that relatively distant LN metastases along the celiac axis and/or the proximal field had a negative impact on survival. Overall, our study emphasized the prognostic importance of the regional location of LN metastases in mHSPC. At present, the CHAARTED and LATITUDE criteria are widely used to define disease burden and risk, respectively. However, these criteria are not universally accepted by clinicians, as they are solely based on the site and number of bone and visceral metastases.17 Our results showed that irrespective of the disease risk or disease burden, patients with concomitant bone and NRLN metastases had similarly poor survival outcomes compared with those harboring visceral metastases, indicating that concomitant bone and NRLN metastases should also be regarded as a risk factor when evaluating disease burden in patients with mHSPC. According to our findings, the metastatic pattern of LN metastases had important prognostic value and should not be ignored in defining disease burden in mHSPC. Finally, we defined G1 and G2 as low risk, G3a and G4 as intermediate risk, and G3b and G5 as high risk. A merged predictive model based on metastatic site was established.
Metastatic patterns could not only facilitate risk classification but also potentially guide treatment decision-making. Several recent studies have suggested that early systemic therapy (DOC, ABI and apalutamide) prolongs OS compared with traditional ADT alone in patients with mHSPC, particularly in those with high risk and a high tumor burden.5,11,18 However, owing to economic concerns, patients’ willingness, and medical insurance policies, few mHSPC patients in China in recent years have received ABI/DOC. In this study, there were relatively more patients receiving ABI/DOC in the G2 and G3 groups, reaching 17 (18.5%) and 16 (16.8%) patients, respectively. We found that there was similar treatment efficacy between MAB and ABI/DOC in the G2 group (CFS: 28.7 months vs 21.4 months, P=0.283; OS: not reached vs 65.5 months, P = 0.266), while ABI/DOC significantly improved CFS and OS compared with MAB in the G3 group (CFS: 28.0 months vs 11.8 months, P = 0.010; OS: not reached vs 42.4 months, P = 0.021; Supplementary Figure 2 (689.9KB, tif) ), which suggested that intensive treatments might be considered the first choice in men with concomitant bone and NRLN metastases. Considering the small number of patients treated with ABI/DOC and their short follow-up time, these results should be interpreted cautiously. We plan to validate our present findings by analyzing the median 5-year follow-up of the patients in the future.
Our study had some limitations. First, this was a retrospective study with a relatively small sample size. Nevertheless, it was the first study to explore the impact of the metastatic patterns of NLRN. We believe that this study has implications for future studies focusing on the metastatic patterns of PCa. Second, LN metastases might have remained undetected in some patients since most patients underwent chest and abdomen enhanced scans, which are less sensitive for detecting LN metastases than positron emission tomography-CT.19
CONCLUSIONS
Metastatic site can predict the prognosis of patients with mHSPC, and the regional location of LN metastases has important prognostic value. Compared to the presence of only bone or concomitant bone and RLN metastases, the presence of concomitant bone and NRLN metastases was a poor prognostic factor in mHSPC patients. Furthermore, based on the LN metastatic plane, we found that the farther the NRLN metastases were located (above or below the diaphragm), the more aggressive the disease was. Nevertheless, further studies on the metastatic patterns of PCa are needed to validate our findings.
AUTHOR CONTRIBUTIONS
ZPW designed this study, analyzed the data, and drafted the manuscript. JRC participated in the statistical analysis and drafted the manuscript. JGZ, GXS, and SZ participated in the study design. XMZ, JYL, BH, and YCN helped with the data collection. PFS and HZ supervised this project and provided funding. All authors read and approved the final manuscript.
COMPETING INTERESTS
All authors declare no competing interests.
Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.
Risk stratification in patients with concomitant bone and NRLN metastases (a and b: CHARRTED; c and d: LATITUDE). NRLN: nonregional lymph nodes; CFS: castration-resistant prostate cancer-free survival; OS: overall survival; mCFS: median CFS; mOS: median OS.
Kaplan‒Meier curves of CFS and OS in the comparison between traditional endocrine therapy and ABI/DOC for men with concomitant bone and LN metastases (a and b: concomitant bone and subdiaphragmatic NRLN metastases; c and d: concomitant bone and supradiaphragmatic NRLN metastases). CFS: castration-resistant prostate cancer-free survival; OS: overall survival; NRLN: nonregional lymph nodes; ABI: abiraterone; DOC: docetaxel; mCFS: median CFS; mOS: median OS.
ACKNOWLEDGMENTS
This work was supported by the National Natural Science Foundation of China (No. 82172785, 82103097, 81974398, 81902577, and 81872107), the Science and Technology Support Program of Sichuan Province (2021YFS0119), and the 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (No. 0040205301E21).
REFERENCES
- 1.Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2021. CA Cancer J Clin. 2021;71:7–33. doi: 10.3322/caac.21654. [DOI] [PubMed] [Google Scholar]
- 2.Zhu Y, Mo M, Wei Y, Wu J, Pan J, et al. Epidemiology and genomics of prostate cancer in Asian men. Nat Rev Urol. 2021;18:282–301. doi: 10.1038/s41585-021-00442-8. [DOI] [PubMed] [Google Scholar]
- 3.Gandaglia G, Karakiewicz PI, Briganti A, Passoni NM, Schiffmann J, et al. Impact of the site of metastases on survival in patients with metastatic prostate cancer. Eur Urol. 2015;68:325–34. doi: 10.1016/j.eururo.2014.07.020. [DOI] [PubMed] [Google Scholar]
- 4.Halabi S, Kelly WK, Ma H, Zhou H, Solomon NC, et al. Meta-analysis evaluating the impact of site of metastasis on overall survival in men with castration-resistant prostate cancer. J Clin Oncol. 2016;34:1652–9. doi: 10.1200/JCO.2015.65.7270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kyriakopoulos CE, Chen YH, Carducci MA, Liu G, Jarrard DF, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer: long-term survival analysis of the randomized phase III E3805 CHAARTED trial. J Clin Oncol. 2018;36:1080–7. doi: 10.1200/JCO.2017.75.3657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tosoian JJ, Gorin MA, Ross AE, Pienta KJ, Tran PT, et al. Oligometastatic prostate cancer: definitions, clinical outcomes, and treatment considerations. Nat Rev Urol. 2017;14:15–25. doi: 10.1038/nrurol.2016.175. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gandaglia G, Abdollah F, Schiffmann J, Trudeau V, Shariat SF, et al. Distribution of metastatic sites in patients with prostate cancer: a population-based analysis. Prostate. 2014;74:210–6. doi: 10.1002/pros.22742. [DOI] [PubMed] [Google Scholar]
- 8.Ali A, Hoyle A, Mistry H, Clarke NW. Importance of non-regional lymph nodes in assigning risk in primary metastatic prostate cancer. BJU Int. 2019;123:65–73. doi: 10.1111/bju.14400. [DOI] [PubMed] [Google Scholar]
- 9.Heesterman BL, van der Poel HG, Schoots IG, Mehra N, Aben KK. Prognostic importance of concomitant non-regional lymph node and bone metastases in men with newly diagnosed metastatic prostate cancer. BJU Int. 2022;130:217–25. doi: 10.1111/bju.15632. [DOI] [PubMed] [Google Scholar]
- 10.Sweeney CJ, Chen YH, Carducci M, Liu G, Jarrard DF, et al. Chemohormonal therapy in metastatic hormone-sensitive prostate cancer. N Engl J Med. 2015;373:737–46. doi: 10.1056/NEJMoa1503747. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Fizazi K, Tran N, Fein L, Matsubara N, Rodriguez-Antolin A, et al. Abiraterone plus prednisone in metastatic, castration-sensitive prostate cancer. N Engl J Med. 2017;377:352–60. doi: 10.1056/NEJMoa1704174. [DOI] [PubMed] [Google Scholar]
- 12.Cornford P, van den Bergh RC, Briers E, Van den Broeck T, Cumberbatch MG, et al. EAU-EANM-ESTRO-ESUR-SIOG guidelines on prostate cancer. Part II-2020 update: treatment of relapsing and metastatic prostate cancer. Eur Urol. 2021;79:263–82. doi: 10.1016/j.eururo.2020.09.046. [DOI] [PubMed] [Google Scholar]
- 13.Heidenreich A, Bastian PJ, Bellmunt J, Bolla M, Joniau S, et al. EAU guidelines on prostate cancer. Part II: treatment of advanced, relapsing, and castration-resistant prostate cancer. Eur Urol. 2014;65:467–79. doi: 10.1016/j.eururo.2013.11.002. [DOI] [PubMed] [Google Scholar]
- 14.Shida H, Ban K, Matsumoto M, Masuda K, Imanari T, et al. Prognostic significance of location of lymph node metastases in colorectal cancer. Dis Colon Rectum. 1992;35:1046–50. doi: 10.1007/BF02252994. [DOI] [PubMed] [Google Scholar]
- 15.de Meer SG, Dauwan M, de Keizer B, Valk GD, Borel Rinkes IH, et al. Not the number but the location of lymph nodes matters for recurrence rate and disease-free survival in patients with differentiated thyroid cancer. World J Surg. 2012;36:1262–7. doi: 10.1007/s00268-012-1427-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Anderegg MC, Lagarde SM, Jagadesham VP, Gisbertz SS, Immanuel A, et al. Prognostic significance of the location of lymph node metastases in patients with adenocarcinoma of the distal esophagus or gastroesophageal junction. Ann Surg. 2016;264:847–53. doi: 10.1097/SLA.0000000000001767. [DOI] [PubMed] [Google Scholar]
- 17.Gillessen S, Attard G, Beer TM, Beltran H, Bjartell A, et al. Management of patients with advanced prostate cancer: report of the advanced prostate cancer consensus conference 2019. Eur Urol. 2020;77:508–47. doi: 10.1016/j.eururo.2020.01.012. [DOI] [PubMed] [Google Scholar]
- 18.Chi KN, Agarwal N, Bjartell A, Chung BH, Pereira de Santana Gomes AJ, et al. Apalutamide for metastatic, castration-sensitive prostate cancer. N Engl J Med. 2019;381:13–24. doi: 10.1056/NEJMoa1903307. [DOI] [PubMed] [Google Scholar]
- 19.Incerti E, Mapelli P, Gianolli L, Picchio M. PET imaging for lymph node dissection in prostate cancer. World J Urol. 2017;35:507–15. doi: 10.1007/s00345-016-1954-8. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Risk stratification in patients with concomitant bone and NRLN metastases (a and b: CHARRTED; c and d: LATITUDE). NRLN: nonregional lymph nodes; CFS: castration-resistant prostate cancer-free survival; OS: overall survival; mCFS: median CFS; mOS: median OS.
Kaplan‒Meier curves of CFS and OS in the comparison between traditional endocrine therapy and ABI/DOC for men with concomitant bone and LN metastases (a and b: concomitant bone and subdiaphragmatic NRLN metastases; c and d: concomitant bone and supradiaphragmatic NRLN metastases). CFS: castration-resistant prostate cancer-free survival; OS: overall survival; NRLN: nonregional lymph nodes; ABI: abiraterone; DOC: docetaxel; mCFS: median CFS; mOS: median OS.