Clinical characteristics of patients. We retrieved 11,966 T1bN0M0 cc-RCC patients who underwent PN or RN between 2004 and 2018 using SEER*Stat 8.3.9 and the extraction process was showed in Figure 1 according to the inclusion and exclusion criteria. Among them, there were greater percentages of <65-year-old patients, males, Caucasians, and tumors measuring 40 mm<T≤50 mm. Regarding all T1bN0M0 cc-RCC patients, their 3-, 5-, and 10-year survival percentages were 91.09%, 84.15%, and 66.54%, respectively, and their median survival was 175 months. Among them, 2,681(22.41%) and 9,285(77.59%) patients underwent PN and RN, respectively. Compared with the RN group, there was a greater percentage of <65-year-old patients, females, and small size tumor cc-RCC patients in the PN group with a significant difference (all P<0.05). The percentages of different racial groups and lateralities were not significantly different for the PN and RN groups. However, there were more patients treated with lymphadenectomy in the RN group than in the PN group (9.13% vs. 4.36%; Table 1).
Survival analysis . Among all patients, the log-rank test indicated that PN patients showed significantly better prognoses than those in the RN group (P<0.0001), with a median survival of 171 months for RN patients and > 179 months for PN patients. After PSM analyses, there were 2681 patients in the PN group and 2,681 in the RN group. After removing the inherent bias of retrospective studies by PSM, survival analysis still showed that the PN group had better prognoses than the RN group (P = 0.0064) (Figure 2).
To further determine the reasons for survival differences, we performed subgroup analyses according to tumor size and age. In terms of tumor size, there were 5768, 3888, and 2310 patients in the 40 mm<T≤50 mm, 50 mm<T≤60 mm, and 60 mm<T≤70 mm groups, respectively. In the 40 mm<T≤50 mm group, there were 1665 and 4103 patients who underwent PN and RN, respectively, and the log-rank tests revealed that PN patients had significantly better prognoses than RN patients (P < 0.0001). However, in the 50 mm<T≤60 mm, and 60 mm<T≤70 mm groups, there was no significant statistical difference between the PN and RN groups. After PSM analysis, 5362 patients remained and there were 3330, 1497, and 535 patients in the 40 mm<T≤50 mm, 50 mm<T≤60 mm, and 60 mm<T≤70 mm groups, respectively. Their clinical information is shown in S Table 2.The log-rank test indicated that in the 40 mm<T≤50 mm group, PN patients had a longer OS (P = 0.0092), but showed comparable results in the 50 mm<T≤60 mm, and 60 mm<T≤70 mm groups (Figure 3).
A subgroup analysis according to age was performed to compare the outcomes of PN and RN at different ages. The patients were divided into <65-year-old group and ≥65-year-old group. The patient characteristics are listed in S Table 3. The log-rank test indicated that in cc-RCC patients <65 years of age, the PN group had a better OS than the RN group (P = 0.0025) and in the patients ≥65 years group patients who underwent PN had a better OS than those who underwent RN (P= 0.0068). After PSM analysis, 3324 patients remained in the < 65 years of age group, while 2038 patients remained in the ≥ 65-year-old group (S Table 4). The log-rank test confirmed that in the two patient age subgroups, PN patients had a longer OS with a significant difference. (Figure 4)
Cox regression analysis. Univariate Cox regression analysis showed that younger age, female sex, other racial groups (including Indian/AK Native, Asian/Pacific Islanders), smaller tumor size, and PN were associated better prognoses, while tumor laterality and lymphadenectomy were not associated with the OS. The factors of age, sex, race, tumor size, and surgery were included into a multivariate Cox proportional hazards model. The results indicated that ≥ 65 years of age , 60 mm<T≤ 70 mm, and RN were associated with a worse survival, and other races (including Indian/AK Native, Asian/Pacific Islanders) had better prognoses (Table 2).