Temporal Patterns and Resolution of Toxicities Following Hypofractionated Salvage Radiotherapy for Biochemical Recurrence of Prostate Cancer After Prostatectomy Authors: Bum‐Sup Jang, Seok‐Joo Chun, Jin Ho Kim Published: 2026 ABSTRACT ABSTRACTBackgroundHypofractionated salvage radiotherapy is increasingly used for biochemical recurrence of prostate cancer post‐prostatectomy, yet its toxicity profile remains underexplored. This study evaluates the incidence, timing, and resolution of treatment‐related toxicities in this setting.MethodsA retrospective cohort study of 403 men receiving hypofractionated salvage radiotherapy (62.5 Gy in 25 fractions) from 2017 to 2024 was conducted. Toxicities, categorized as genitourinary (hematuria, urinary incontinence, frequency, dysuria) or gastrointestinal (rectal bleeding, diarrhea, proctitis, nausea), were graded (1–3) using CTCAE‐based criteria. Outcomes included any‐time incidence, Kaplan–Meier time‐to‐first‐event estimates (12‐ and 24‐month cumulative incidence), and first‐onset timing. Episode‐level analyses merged events ≤ 30 days apart, assessing resolution and intervention.ResultsMedian follow‐up was 18.2 months. Any‐grade genitourinary and gastrointestinal toxicities occurred in 34.2% and 24.6% of patients, respectively (24‐month incidence: genitourinary 40.4%; gastrointestinal 28.9%). Grade 2+ toxicities were less frequent (genitourinary 9.7%; gastrointestinal 7.7%), with Grade 3+ rare (genitourinary 1.7%; gastrointestinal 0.7%). In terms of patient‐level ever‐event incidences, common symptoms included urinary incontinence (20.3%), rectal bleeding (19.9%), and hematuria (13.9%). Urinary toxicities predominantly onset at 6–12 months, whereas rectal bleeding and hematuria emerged later (≥ 12 months). Approximately two‐thirds of Grade 2+ episodes resolved (e.g., urinary incontinence, 66.7%; rectal bleeding, 69.2%), with median resolution times of 3.0–8.4 months.ConclusionWith a median follow‐up of 18.2 months, hypofractionated salvage radiotherapy demonstrates favorable short‐to‐intermediate term tolerability, with infrequent severe toxicities. Vigilant monitoring for late‐onset rectal bleeding and hematuria is recommended during the 6–24 months after treatment to optimize patient management. Longer follow‐up is required to confirm long‐term safety. Introduction Prostate cancer is one of the most common malignancies among men worldwide [1]. Despite advances in the treatment of prostate cancer, biochemical recurrence (BCR) following radical prostatectomy remains a major clinical challenge, occurring in approximately 20%–40% of patients [2, 3]. BCR, typically defined as a rise in prostate‐specific antigen after surgery, often precedes radiographic evidence of disease progression and may serve as an early indicator of systemic relapse [4]. Effective management of BCR is therefore essential, as timely intervention can modify the disease course, delay metastasis, and ultimately improve survival outcomes. Salvage radiotherapy (RT) has emerged as the standard of care for patients with BCR after prostatectomy, with numerous studies demonstrating its potential to provide durable disease control when appropriately administered [5, 6, 7, 8]. Traditionally, salvage RT has been delivered using conventional fractionation schedules, with doses of 1.8–2.0 Gy over 7 weeks. While this approach is well established and effective, it is associated with financial burdens for patients and healthcare systems, as well as the potential for prolonged exposure to treatment‐related toxicities. In recent years, hypofractionated RT, characterized by larger fraction sizes and a shorter overall treatment course, has emerged as an attractive alternative [9, 10, 11]. The rationale for hypofractionation is supported by the unique radiobiological characteristics of prostate cancer, particularly its low alpha/beta ratio [12, 13], which suggests greater sensitivity to a larger dose per fraction. Moreover, hypofractionation has the potential to improve patient convenience, reduce healthcare costs, and increase accessibility without compromising efficacy. Although hypofractionated RT has demonstrated efficacy in the salvage setting, concerns remain regarding its safety and tolerability. Radiation‐induced toxicities, particularly genitourinary (GU) and gastrointestinal (GI) complications, represent a major determinant of treatment feasibility and patient quality of life. While multiple prospective trials have established the safety of hypofractionation for definitive treatment of localized prostate cancer, evidence supporting its use in salvage therapy remains limited. Patients with BCR often have a history of prior surgery or radiation, which may predispose them to heightened sensitivity and increased risk of toxicity. Consequently, a thorough evaluation of treatment‐related adverse effects in this population is essential. The present study aims to evaluate the toxicity outcomes of hypofractionated RT in patients with BCR of prostate cancer. Ultimately, elucidating the tolerability of hypofractionated RT, including pelvic nodal irradiation, is critical for optimizing treatment strategies and ensuring high‐quality patient care in this complex clinical context. Materials and Methods Study Design and Cohort Characteristics We retrospectively analyzed post‐RT toxicities using a de‐identified data set linking event‐level toxicities to patient‐level variables. Eligible patients underwent salvage intensity‐modulated RT (IMRT) between 2017 and 2024 after BCR following radical prostatectomy. Treatment consisted of 62.5 Gy in 25 fractions to the prostate fossa, with concurrent elective pelvic nodal irradiation (presacral, common, internal, and external iliac nodes) to 45 Gy in 25 fractions delivered via a simultaneous integrated boost. Ethical approval was waived by the Institutional Review Board owing to the use of a fully de‐identified data set. Data Harmonization and Variable Definitions Patient‐level variables included age, treatment dates, Gleason score, pathologic stage (parsed T/N/M), and concurrent androgen‐deprivation therapy (ADT) with standardized agents. Event‐level data encompassed symptom, grade, intervention, onset and resolution dates or months. Toxicities were evaluated based on physician physical examinations and patient‐reported symptoms documented in the electronic medical records during routine clinic visits, which were then retrospectively graded using the CTCAE criteria. Analyses focused on eight predefined toxicities: four GU (hematuria, urinary incontinence, urinary frequency, dysuria) and four GI (rectal bleeding, diarrhea, proctitis, nausea). Specifically, “rectal bleeding” was recorded based on clinical symptoms, whereas “proctitis” strictly required a formal clinical or endoscopic diagnosis of mucosal inflammation. Toxicities were graded from 1 to 5 according to the CTCAE criteria, with Grade ≥ 2 and Grade ≥ 3 toxicities grouped as “2+” and “3+,” respectively. Interventions were categorized hierarchically: Supportive/None, Medication, Procedure, and Transfusion/Surgery/Hospitalization (TSH), with the highest‐intensity category assigned per episode. Time Origin, Follow‐Up, and Episode Construction The time origin (t = 0) was the RT end date. Follow‐up was defined from RT end to last documented visit; missing follow‐up was imputed from observed onset/resolution. Events within 30 days for the same patient and symptom were merged into a single episode, defined by earliest onset and latest resolution, and assigned maximum grade and primary intervention. Episodes were classified as resolved if recovery was documented before censoring. Endpoints and Statistical Analysis Primary endpoints were: (1) cumulative incidence of first toxicity at any grade, Grade 2+, and Grade 3+ using Kaplan–Meier methods with Greenwood 95% confidence intervals; (2) ever‐event proportions with Wilson intervals; (3) timing distributions by grade within 0–3, 3–6, 6–12, 12–24, and > 24 months; (4) point prevalence of active toxicities at 0.5‐month intervals; and (5) management patterns by intervention category. All analyses were conducted in Python using Pandas, NumPy, SciPy, Lifelines, Seaborn, Matplotlib, Plotly, and Kaleido. Kaplan–Meier estimates were computed with lifelines, Wilson intervals for binomial proportions, Savitzky–Golay smoothing via scipy, and Sankey diagrams via plotly, exported as static images. Results Study Population A total of 403 patients were included. Median age at RT was 69.0 years (IQR 64.0–74.0); median follow‐up was 18.2 months (IQR 6.7–30.1). ADT was administered during RT in 74.4% of cases. Most patients had pT3 disease (66.0%); nodal staging was pN0 in 39.0% (Table S1). Patient‐Level Incidence and Timing Across individual symptoms, any‐grade ever‐event incidence was highest for urinary incontinence (20.3%; 95% CI, 16.7–24.5), rectal bleeding (19.9%; 95% CI, 16.2–24.0), and hematuria (13.9%; 95% CI, 10.9–17.6), followed by urinary frequency (7.9%; 95% CI, 5.7–11.0), diarrhea (5.2%; 95% CI, 3.4–7.8), proctitis (5.0%; 95% CI, 3.2–7.5), dysuria (4.2%; 95% CI, 2.7–6.7), and nausea (2.7%; 95% CI, 1.5–4.8). Grade 2+ ever‐event incidence was 6.7% (95% CI, 4.6–9.6) for rectal bleeding, 6.0% (95% CI, 4.0–8.7) for urinary incontinence, 4.7% (95% CI, 3.0–7.2) for hematuria, 3.0% (95% CI, 1.7–5.1) for proctitis, 0.7% (95% CI, 0.3–2.2) for urinary frequency, and 0.5% (95% CI, 0.1–1.8) for diarrhea; no Grade 2+ dysuria or nausea occurred (Table 1). Table 1Incidence and timing summary by symptom (patient‑level).CategoryVariableUrinary incontinenceHematuriaUrinary frequencyDysuriaRectal bleedingDiarrheaProctitisNauseaAny gradeEver‐event Incidence (95% CI)20.3 [16.7, 24.5]13.9 [10.9, 17.6]7.9 [5.7, 11.0]4.2 [2.7, 6.7]19.9 [16.2, 24.0]5.2 [3.4, 7.8]5.0 [3.2, 7.5]2.7 [1.5, 4.8]Cumulative Incidence at 12 months15.8 [12.4, 20.1]3.7 [2.1, 6.5]7.6 [5.3, 10.9]3.9 [2.3, 6.6]14.4 [11.0, 18.7]5.3 [3.4, 8.3]3.3 [1.8, 6.0]2.8 [1.5, 5.2]Cumulative incidence at 24 months26.4 [21.5, 32.2]18.7 [13.9, 24.9]9.1 [6.4, 12.9]4.9 [3.0, 8.1]24.9 [20.1, 30.8]5.3 [3.4, 8.3]6.7 [4.2, 10.6]3.2 [1.8, 5.7]Grade 2 or higherEver‐event incidence (95% CI)6.0 [4.0, 8.7]4.7 [3.0, 7.2]0.7 [0.3, 2.2]0.0 [0.0, 0.9]6.7 [4.6, 9.6]0.5 [0.1, 1.8]3.0 [1.7, 5.1]0.0 [0.0, 0.9]Cumulative incidence at 12 months4.0 [2.4, 6.7]1.0 [0.3, 3.0]0.8 [0.3, 2.6]0.0 [0.0, 0.0]6.1 [3.9, 9.4]0.3 [0.0, 2.1]2.3 [1.1, 4.8]0.0 [0.0, 0.0]Cumulative incidence at 24 months8.8 [5.8, 13.1]6.4 [3.7, 10.8]0.8 [0.3, 2.6]0.0 [0.0, 0.0]9.1 [6.2, 13.4]0.3 [0.0, 2.1]3.8 [2.1, 6.7]0.0 [0.0, 0.0]First onset (any grade)Median time to first onset in months [IQR]6.1 [6.0, 17.9]17.8 [13.8, 23.8]6.0 [5.5, 6.2]6.0 [5.7, 6.4]9.3 [5.5, 15.9]6.0 [3.0, 6.2]11.8 [8.0, 16.6]6.0 [2.0, 6.5]First onset in 6–12 months (%)42.716.15047.132.542.94054.5First onset in 12–24 months percent (%)29.353.69.411.8300409.1Abbreviations: CI, confidence interval; IQR, interquartile range; mo, months; RT, radiation therapy.John Wiley & Sons, Ltd. Incidence and timing summary by symptom (patient‑level). Abbreviations: CI, confidence interval; IQR, interquartile range; mo, months; RT, radiation therapy. In total, 12‐ and 24‐month cumulative incidence (any grade) were 15.8% (95% CI, 12.4–20.1) and 26.4% (95% CI, 21.5–32.2) for urinary incontinence; 14.4% (95% CI, 11.0–18.7) and 24.9% (95% CI, 20.1–30.8) for rectal bleeding; and 3.7% (95% CI, 2.1–6.5) and 18.7% (95% CI, 13.9–24.9) for hematuria, respectively. For Grade 2+ thresholds, 24‐month cumulative incidence was 9.1% (95% CI, 6.2–13.4) for rectal bleeding, 8.8% (95% CI, 5.8–13.1) for urinary incontinence, 6.4% (95% CI, 3.7–10.8) for hematuria, 3.8% (95% CI, 2.1–6.7) for proctitis, and 0.8% (95% CI, 0.3–2.6) for urinary frequency; diarrhea plateaued by 12 months (0.3% at 12 and 24 months; 95% CI, 0.0–2.1) (Table 1 and Figure 1A,B). Time dynamics of post‑RT toxicity by organ system. (A) GU cumulative incidence of first toxicity by symptom and grade threshold. Colors denote symptoms; line styles denote thresholds (any grade, Grade 2+, Grade 3+). X‑axis, months since RT end (0–60, 12‑month ticks); Y‑axis, cumulative incidence defined as 1 − Kaplan–Meier S(t). (B) GI cumulative incidence, displayed as in (A). (C) GU point prevalence of active episodes over time. Prevalence is the number of unique patients with an active episode at time t divided by the total cohort size; curves are smoothed by a Savitzky–Golay filter. The Y‑axis is fixed to 0–0.20. (D) GI point prevalence, displayed as in (C). RT, radiation therapy; KM, Kaplan–Meier; GU, genitourinary; GI, gastrointestinal. [Color figure can be viewed at wileyonlinelibrary.com] The median time to first onset (any grade) illustrated distinct temporal patterns: hematuria was late (17.8 months; IQR, 13.8–23.8), rectal bleeding intermediate (9.3 months; IQR, 5.5–15.9), and urinary incontinence early‐to‐intermediate (6.1 months; IQR, 6.0–17.9). Urinary frequency, dysuria, diarrhea, and nausea clustered around 6 months, whereas proctitis was later (11.8 months; IQR, 8.0–16.6). Consistently, first onsets occurred most often at 6–12 months for urinary symptoms (urinary incontinence 42.7%; urinary frequency 50.0%; dysuria 47.1%) and at ≥ 12 months for hematuria (12–24 months, 53.6%) and proctitis (12–24 months, 40.0%). For rectal bleeding, any‐grade first onset was distributed across 6–12 months (32.5%) and 12–24 months (30.0%) (Table 1). Complementing these first‑onset findings, the smoothed point prevalence of active episodes over time (Figure 1C,D) remained low on the fixed 0–0.20 scale across both GU and GI panels. Modest, time‑limited rises in prevalence align with the predominant first‑onset windows—earlier for urinary incontinence and urinary frequency (around 6–12 months) and later for hematuria, rectal bleeding, and proctitis (around 12–24 months)—and then recede, consistent with the observed episode resolution rates and the rarity of patient‑level recurrences. Grade‐Stratified Timing Heatmaps stratified by grade (Figure 2) illustrate the percentage of the entire study cohort with first onset in predefined time windows. For Grade 1 events, first onsets were most prominent at 6–12 months for urinary incontinence (7.2%), rectal bleeding (3.0%), urinary frequency (3.7%), and dysuria (2.0%). Grade 1 hematuria was predominantly late, with 4.7% at 12–24 months and 2.5% at > 24 months. For Grade 2 events, rectal bleeding peaked at 6–12 months (3.7% of cohort), urinary incontinence at 12–24 months (2.7%), and proctitis at 6–12 months (1.2%); hematuria Grade 2 was later (1.7% at 12–24 months and 1.0% at > 24 months). Grade 3+ events were rare overall, with hematuria showing scattered first onsets (0.5% at > 24 months, 0.2% at 3–6 months and 1.0% at 12–24 months), and rectal bleeding limited to 0.2% at 12–24 and > 24 months. No Grade 3+ events were observed for urinary incontinence, urinary frequency, dysuria, diarrhea, or nausea (Figure S1 and Table S2). Grade‑stratified first‑onset heatmaps. (A) Grade 1. Y‑axis, symptom; X‑axis, first‑onset windows (0–3, 3–6, 6–12, 12–24, > 24+ months). Each cell displays the percentage of the entire cohort with a first onset in that window; independent color scales are used per panel. (B) Grade 2, displayed as in (A). (C) Grade 3+, displayed as in (A). %, percent. [Color figure can be viewed at wileyonlinelibrary.com] Management, Resolution, and Persistence of Symptoms At the episode level, management varied (Figure 3A,B and Table 2). Among Grade 2 episodes, procedures were frequently used for GI toxicities—rectal bleeding (57.7%) and proctitis (50.0%)—whereas Grade 2 urinary incontinence was managed predominantly with supportive measures or no specific intervention (91.7%). All Grade 2 diarrhea episodes were managed supportively (100%). For Grade 3+ episodes, escalation of care occurred in a subset: hematuria required procedures in 42.9% and TSH in 14.3%; rectal bleeding episodes were extremely rare (n = 2) and involved medication (mesalazine) in 50% and transfusion in 50%. For Grade 2 rectal bleeding, procedures consisted predominantly of argon plasma coagulation, with mesalazine suppositories used as pharmacologic management. The single Grade 3+ proctitis episode underwent a procedure (100%). An episode‐level Sankey diagram maps the flow from symptom to grade and then to the primary intervention (Figure 3C). Across Grade 2 episodes, resolution during follow‐up occurred in 69.2% of hematuria episodes (median time to resolution, 3.0 months), 69.2% of rectal bleeding (8.4 months), 66.7% of urinary incontinence (5.0 months), 100% of urinary frequency (6.3 months), 75.0% of proctitis (8.1 months), and 50.0% of diarrhea (0.0 months). Among Grade 3+ episodes, resolution was observed in 71.4% of hematuria (3.6 months), 50.0% of rectal bleeding (11.3 months), and 100% of proctitis (12.2 months) (Table 2 and Figure 4A,B). A subset of episodes, particularly rectal bleeding and hematuria, persisted beyond 12 months, whereas dysuria, urinary frequency, and diarrhea typically resolved within 3 months (Figure 4B and Table S3). Recurrence was uncommon, with over 90% of patients experiencing only a single episode per symptom (Figure 4C and Table S4). Interventions and escalation of care. (A) Grade 2 primary intervention category by symptom, shown as 100% stacked bars. Categories are Supportive/None, Medication, Procedure, and Transfusion/Surgery/Hospitalization. (B) Grade 3+ primary intervention category by symptom, displayed as in (A). (C) Sankey diagram depicting episode‑level flows from Symptom to Grade to Intervention. Node labels show “name (count, % of all episodes)”; link labels show “count (within‑source %).” Supportive/None, supportive measures or no documented intervention; TSH, Transfusion/Surgery/Hospitalization. [Color figure can be viewed at wileyonlinelibrary.com] Table 2Management and outcomes by symptom and grade (episode‑level; Grades 2 and 3+).GradeSymptomHematuriaUrinary incontinenceUrinary frequencyRectal bleedingDiarrheaProctitisGrade 2Supportive/none69.291.766.738.510025Medication08.333.33.8025Procedure30.80057.7050Transfusion, surgery, or hospitalization000000Resolved69.266.710069.25075Median time to resolution in months [IQR]3.0 [2.0, 7.0]5.0 [0.0, 12.2]6.3 [3.2, 6.8]8.4 [4.5, 15.0]0.0 [0.0, 0.0]8.1 [4.4, 12.2]Grade 3Supportive/none42.9N/AN/A0N/A0Medication0N/AN/A50N/A0Procedure42.9N/AN/A0N/A100Transfusion, surgery, or hospitalization14.3N/AN/A50N/A0Resolved71.4N/AN/A50N/A100Median time to resolution in months [IQR]3.6 [1.5, 12.3]N/AN/A11.3 [11.3, 11.3]N/A12.2 [12.2, 12.2]RecurrencePatients with two or more episodes (%)7.14.93.1104.810Abbreviations: IQR, interquartile range; %, percent; N/A, not available; Supportive/none, supportive care or no documented intervention.John Wiley & Sons, Ltd. Management and outcomes by symptom and grade (episode‑level; Grades 2 and 3+). Abbreviations: IQR, interquartile range; %, percent; N/A, not available; Supportive/none, supportive care or no documented intervention. Clinical course summary. (A) Resolved episodes percent by symptom for Grades 2 and Grade 3+, shown as side‑by‑side bars. Resolution is defined as observed resolution before patient‑level censoring, optionally within a prespecified horizon if used. (B) Time‑to‑resolution distributions among resolved episodes, shown as violin plots with inner quartiles or as dots with IQR whiskers (Grade 2 vs. Grade 3+). Symptoms with no data are omitted. (C) Patient‑level recurrence, showing the percentage of patients with two or more episodes by symptom. Symptoms with no data are omitted. IQR, interquartile range; ≥, greater than or equal to. [Color figure can be viewed at wileyonlinelibrary.com] Taken together, these data demonstrate that most toxicities first emerge between 6 and 24 months after radiotherapy, with later and more protracted courses for hematuria and rectal bleeding, limited need for escalation of care outside GI events, and resolution of approximately two‐thirds of Grade 2+ episodes within months. Discussion Hypofractionated RT is established as a standard for localized prostate cancer, with trials such as CHHiP, PROFIT, and HYPRO confirming noninferior efficacy and acceptable safety [9, 10, 11]. Concerns persist about bladder and rectal sensitivity to larger fraction sizes, particularly in the postoperative setting. In this retrospective cohort of 403 patients receiving hypofractionated salvage RT (62.5 Gy/25 fractions) with concurrent pelvic nodal irradiation, overall toxicity was low: any‐grade urinary incontinence, rectal bleeding, and hematuria occurred in 20.3%, 19.9%, and 13.9% of patients, while Grade 2+ events were uncommon and Grade 3+ events rare. Most toxicities resolved with conservative care, with escalated interventions required only for a minority of severe cases. Our findings align closely with two Phase 3 trials directly comparing hypofractionated and conventional salvage RT [14, 15]. In the NRG‐GU003 trial (n = 296), which randomized patients to hypofractionated RT (HYPORT: 62.5 Gy in 25 fractions) versus conventional RT (COPORT: 66.6 Gy in 37 fractions), HYPORT was noninferior for patient‐reported late GU and GI toxicities at 2 years, with no significant increase in Grade 2+ events (GU: ~10%–15%; GI: 5%–10%, based on Expanded Prostate Cancer Index Composite [EPIC] scores) [14]. Notably, while acute GI symptoms were modestly higher in HYPORT at RT completion, late severe toxicities (Grade 3+) remained low (< 5%) in both arms, mirroring our cohort's Grade 3+ GI rate of 0.7%. Similarly, our any‐grade GI incidence (24.6%) and Grade 2+ rate (7.7%) are comparable to NRG‐GU003's long‐term data, where compliance with bowel and urinary EPIC domains exceeded 80% at 2 years, suggesting preserved quality of life. Thus, our hypofractionated regimen, even with pelvic nodal inclusion, does not appreciably elevate toxicity beyond that observed in NRG‐GU003, where nodal fields were not routinely irradiated. The RADICALS‐RT trial (n = 1396), evaluating adjuvant or salvage RT with hypofractionated (52.5 Gy in 20 fractions) versus conventional (66 Gy in 33 fractions) schedules, further corroborates our results [15]. Toxicity profiles were broadly comparable between the two groups, with no significant differences in late GI or GU adverse events. A slight increase in Grades 1–2 cystitis was observed in the conventional arm (25% vs. 20% in hypofractionated), but this did not translate into clinically meaningful higher‐grade toxicity, with Grade 3+ events occurring in < 2% across arms. Other toxicities, including rectal bleeding and urinary incontinence, were similar and infrequent (Grade 2+ rectal bleeding ~5%–7%; urinary incontinence ~10%–12%). Patients reported slightly higher urinary symptom scores in the hypofractionated group at 1 year, but these differences resolved by 5 years. In comparison, our cohort exhibited comparable or slightly higher any‐grade rectal bleeding (19.9% vs. ~15%–20% in RADICALS‐RT) and urinary incontinence (20.3% vs. ~15%), but Grade 2+ rates remained low and consistent (6.7% for rectal bleeding; 6.0% for urinary incontinence). These minor differences may reflect our routine inclusion of pelvic nodal irradiation, which could contribute to modestly elevated late GI events, though our Grade 3+ rates (< 1%) mirror the rarity observed in RADICALS‐RT. In our cohort, pelvic nodal regions were all included in the radiation field, with 45 Gy delivered to the presacral, common iliac, internal iliac, and external iliac nodes. This safety profile is further supported by the SPPORT trial (RTOG 0534; n = 1792), which demonstrated that adding pelvic lymph node RT to prostate bed RT (with or without short‐term ADT) improved progression‐free survival without substantial increases in late toxicity [16]. Although pelvic field inclusion was associated with a modest increase in acute GI toxicities (Grade 2+: 18% with pelvic nodes vs. 12% without), differences in late toxicity were minimal (late Grade 2+ GI: 10%–15%; GU: 15%–20%; Grade 3+ < 3%). Our observation of low late‐grade GI events (7.7% Grade 2+; 0.7% Grade 3+) is consistent with this evidence, supporting the safety of incorporating pelvic irradiation into hypofractionated regimens. Furthermore, when compared with reported toxicities from the RADICALS‐RT and NRG‐GU003 trials [14, 15], which did not routinely include pelvic lymph nodes, our data suggest that elective pelvic nodal irradiation can be delivered with acceptable safety and without a disproportionate increase in long‐term toxicity. Additionally, recent systematic reviews of hypofractionated postoperative/salvage RT (encompassing > 10 studies, including NRG‐GU003 and RADICALS‐RT) showed moderate toxicity rates (late GU Grade 2+: 5%–15%; GI: 5%–10%) and noninferiority to conventional regimens, with biochemical control rates > 80% at 5 years [17, 18]. The temporal distribution of toxicities revealed distinct patterns between GU and GI domains. Urinary toxicities, such as incontinence and frequency, occurred predominantly within the first 6–12 months after radiotherapy, whereas GI toxicities, particularly rectal bleeding, were more prominent during the 12 to 24‐month interval. This chronology is consistent with the biological mechanisms underlying radiation injury. Early urinary toxicities are often attributable to mucosal inflammation and transient functional changes in the bladder outlet or urethra, which typically manifest within the first year and improve with conservative management [19]. In contrast, rectal bleeding tends to arise later, reflecting progressive vascular and mucosal changes such as telangiectasia and chronic ischemia, which characteristically emerge 1–2 years after treatment [20, 21]. Similar latency patterns have been reported in prospective trials of postoperative radiotherapy, including RADICALS‐RT and NRG‐GU003, where urinary irritative symptoms peaked earlier (e.g., Grade 2+ urinary frequency ~5%–10% at 6–12 months) while late GI bleeding accumulated over longer follow‐up (e.g., Grade 2+ rectal bleeding ~5%–8% at 12–24 months) [14, 15]. Our episode‐level resolution data further support this, with two‐thirds of Grade 2+ episodes resolving within 3–8 months, aligning with resolution rates of 60%–80% in NRG‐GU003 and SPPORT [14, 16]. In terms of pelvic nodal irradiation, SPPORT [16] demonstrated GI events were predominantly acute, with late prevalence < 5%, supporting our smoothed prevalence curves (fixed at < 20% active episodes). These insights advocate for risk‐stratified surveillance: intensive GU monitoring in the first year, followed by extended GI follow‐up to 24 months, particularly in patients with pelvic fields. However, it is crucial to recognize that surveillance extending well beyond this 24‐month window is necessary to fully capture late‐onset events. A recent report by Ranta and colleagues with a 13‐year follow‐up reinforced this, demonstrating that while few toxicities occurred within the first 2 years, the median time to severe (Grade 3+) toxicity was 106 months [22]. Importantly, while very late severe events may emerge beyond 5–10 years, several studies indicate that the majority of clinically relevant Grade 2+ toxicities in postoperative radiotherapy have historically manifested within the first 2–3 years [15, 23, 24], suggesting that our findings remain meaningful for early safety assessment. Limitations include the retrospective design, potential underreporting of transient toxicities due to 6‐month follow‐up intervals, and a relatively short median follow‐up of 18.2 months, which may underestimate very late events. Nonetheless, this is among the largest single‐center analyses of hypofractionated salvage RT with pelvic lymph node radiotherapy, providing granular episode‐level data absent in many trials. Future prospective studies should incorporate patient‐reported outcomes (e.g., EPIC) and longer follow‐up to validate these patterns. Conclusion In conclusion, hypofractionated salvage RT with pelvic nodal irradiation yields low short‐to‐intermediate term toxicity rates that are consistent with published phase data, with most events mild and resolvable. These findings support the short‐to‐intermediate term safety of hypofractionated salvage RT with pelvic nodal irradiation and may inform clinical decision‐making, although longer follow‐up is required before definitive conclusions regarding long‐term safety can be drawn. Funding The authors have nothing to report. Ethics Statement Ethics approval was waived due to the use of a fully de‐identified data set. Consent The requirement for informed consent was waived due to the use of a fully de‐identified data set. Conflicts of Interest The authors declare no conflicts of interest. Supporting information Supporting File