摘要目的:评估 68Ga标记的前列腺特异性膜抗原( 68Ga-PSMA)PET/CT对前列腺癌的诊断效能，并探讨 68Ga-PSMA PET/CT对术前制订保留血管神经束(NVB)和淋巴结清扫策略的指导作用。 方法:回顾性分析2018年6月至2019年10月中国医学科学院肿瘤医院行 68Ga-PSMA PET/CT检查的46例初诊疑似前列腺癌患者的临床资料。中位年龄66.50(60.00，69.25)岁，中位前列腺特异性抗原(PSA)值15.97(8.58，33.10)ng/ml。46例中，41例 68Ga-PSMA PET/CT检查诊断为肿瘤，6例诊断有淋巴结转移；5例诊断为前列腺增生或前列腺炎。46例中40例同期行mpMRI检查，33例诊断为肿瘤，6例诊断有淋巴结转移；46例中17例同期行 11C-胆碱PET/CT检查，12例诊断为肿瘤，4例诊断有淋巴结转移。41例PSMA-PET/CT诊断为前列腺癌的患者中，高危22例，中危19例；其中37例行mpMRI检查，15例行 11C-胆碱PET/CT检查。41例均行根治性前列腺切除术。根据 68Ga-PSMA PET/CT显示的肿瘤位置，术前制订NVB处理策略：若肿瘤邻近前列腺单侧包膜，则保留健侧的NVB；若肿瘤局限于前列腺内，则保留双侧NVB。共16例保留了NVB(单侧6例，双侧10例)。对中高危组患者常规行淋巴结清扫。采用配对 χ2检验或Fisher精确检验比较 68Ga-PSMA PET/CT、mpMRI、 11C-胆碱PET/CT对病灶检出的敏感性和特异性。采用Spearman相关分析检测 68Ga-PSMA PET/CT的SUV max值与Gleason评分和治疗前PSA值的相关性。 结果:41例行根治术患者术后病理确诊为前列腺癌，手术切缘均未见癌组织；中位Gleason评分8(7，9)分；病理分期20例≤pT 2c期，21例≥pT 3期；7例淋巴结阳性(11枚阳性淋巴结)。术后30 d内7例(17.1%)发生并发症，Clavien-Dindo分级均≤2级。41例术后随访中位时间16(12，20)个月，术后1、6、12个月分别有19例(46.3%)、39例(95.1%)、41例(100.0%)恢复控尿。5例未行手术的患者中，4例行抗生素治疗后PSA下降；1例PSA未下降者行穿刺活检，病理未见癌。 68Ga-PSMA PET/CT诊断前列腺癌的敏感性为100.0%(41/41)，显著优于 11C-胆碱PET/CT[80.0%(12/15)， P＝0.016]和mpMRI[83.7%(31/37)， P＝0.009]；特异性为100.0%(5/5)，与 11C-胆碱PET/CT[100.0%(2/2), P＝1.000]和mpMRI [33.3%(1/3)， P＝0.107]的差异均无统计学意义。41例中， 68Ga-PSMA PET/CT诊断淋巴结转移的敏感性[71.4%(5/7)]与 11C-胆碱PET/CT的差异无统计学意义[75.0%(3/4)， P＝1.000]，与mpMRI的差异有统计学意义[16.7%(1/6)， P＝0.016]。Gleason评分≥8分与<8分患者 68Ga-PSMA PET/CT的原发灶SUV max值分别为19.60(9.58，24.38)与8.55(5.18，12.88)；治疗前PSA值≥20 ng/ml与<20 ng/ml患者的SUV max值分别为19.40(13.00，23.5)与8.40(5.35，13.95)，差异均有统计学意义( P<0.05)。 结论:68Ga-PSMA PET/CT对前列腺癌原发病灶诊断的敏感性高、特异性高，术前可根据PSMA PET/CT显示的肿瘤位置，制订是否保留NVB的处理策略；但其对淋巴结转移灶诊断的敏感性还不足以指导术前制订淋巴结清扫策略。

abstractsObjective:To evaluate the value of 68Ga-labeled prostate-specific membrane antigen( 68Ga-PSMA PET/CT)in the diagnosis of prostate cancer(PCa) and determination of the strategy on neurovascular bundle(NVB)preservation and lymphadenectomy before surgery. Methods:We retrospectively analyzed the clinical data of 46 patients with newly diagnosed PCa who underwent 68Ga-PSMA PET/CT from June 2018 to October 2019. The median age was 66.50 (60.00, 69.25) years old and the median PSA was 15.97(8.58, 33.10)ng/ml. Forty-one patients were diagnosed PCa and 5 were diagnosed with benign prostatic hyperplasia or prostatitis by 68Ga-PSMA PET/CT, 6 were diagnosed with lymph nodes metastasis. Forty patients underwent mpMRI, 33 were diagnosed PCa, and 6 were diagnosed with lymph nodes metastasis. Seventeen patients underwent 11C-choline PET/CT, 12 were diagnosed PCa, and 4 were diagnosed with lymph nodes metastasis. Among the 41 patients which were diagnosed PCa by 68Ga-PSMA PET/CT, 26 were confirmed with PCa by needling biopsy, 12 did not undergo the needling biopsy and 3 had negative biopsy; 22 were in the high-risk group, and 19 were in the low- and medium-risk group. All 41 patients underwent radical prostatectomy. The strategy of NVB preservation was determined by the position of the PCa reported by 68Ga-PSMA PET/CT. If the tumor was close to the unilateral prostate capsule, the healthy part of the NVB would be preserved. And if the tumor was limited in the prostate gland, bilateral NVB would be preserved. The NVB was preserved in 16 cases(6 cases unilateral and 10 cases bilateral). In addition to routine lymph node dissection for patients in the high-risk group, lymph node dissection was also performed on patients with 68Ga-PSMA PET/CT that showed positive lymph nodes in the low- and medium-risk groups. The paired chi-square or Fisher exact test was used to compare the sensitivity and specificity of 68Ga-PSMA PET/CT, mpMRI, and 11C-choline PET/CT for lesion detection. Spearman analysis was used to examine the correlation between the SUV max, Gleason score, and the PSA value before treatment. Results:Forty-one patients undergoing radical resection were diagnosed PCa by postoperative pathology, and no cancer tissue was observed at the surgical margin; the median Gleason score was 8(7, 9); 20 cases (48.8%) had a pathological stage ≤pT 2c; 21 cases (51.2%)≥pT 3; 7 cases were N+ (11 positive lymph nodes). Seven cases (17.1%) had complications of Clavien-Dindo ≤ grade 2 within 30 days after surgery, and there was no complication above grade 3 after surgery. The median follow-up time of the 41 patients was 16(12, 20). The rate of urinary control was 46.3%, 95.1%, and 100% after 1 month, 6 months, and 12 months follow-up respectively. Among the five patients that did not undergo surgery, the PSA value of 4 decreased after antibiotic treatment, and biopsy was performed in 1 case without PSA decreasing, and no carcinoma was found. The sensitivity of 68Ga-PSMA PET/CT on the primary PCa was 100%(41/41), which was significantly better than that of 11C-choline(80%, 12/15, P=0.016)and mpMRI (83.7%, 31/37, P=0.009), while no statistical significance was observed between the specificity of 68Ga-PSMA PET/CT (5/5)and 11C-choline PET/CT(2/2, P=1.000), 68Ga-PSMA PET/CT and mpMRI(1/3, P=0.107). Of the 41 patients that were diagnosed with PCa, the sensitivity of 68Ga-PSMA PET/CT to lymph nodes metastases(71.4%, 5/7) had a significant difference with that of mpMRI(16.7, 1/6, P=0.016), but no statistical difference with 11C-choline (75%, 3/4, P=1.000). Analysis of the relationship between SUV max of 68Ga-PSMA PET/CT, Gleason score, and PSA value before treatment revealed that the SUV max of 68Ga-PSMA PET/CT in patients with Gleason score ≥8 and <8 score were 19.60(9.58, 24.38) and 8.55 (5.18, 12.88); SUV max of patients with PSA values ≥20 ng/ml and <20 ng/ml before treatment were 19.40 (13.00, 23.50) and 8.40 (5.35, 13.95), respectively, the differences were statistically significant (all P<0.05). Conclusions:68Ga-PSMA PET/CT had high sensitivity and specificity for the diagnosis of primary prostate cancer lesions, but the sensitivity for the diagnosis of lymph node metastasis was not enough to guide the preoperative decision of whether to remove the lymph node and the scope of the removal. However, the treatment strategy of whether to retain NVB could be formulated according to the tumor location displayed by 68Ga-PSMA PET/CT before surgery.