Population-based registry analysis of cervical cancer trends in Tashkent, Uzbekistan 2014-2024 Authors: Ulugbek Khudayberdievich Mirzaev, Jasur Ravshanovich Juraev, Iroda Pulatovna Sharipova, Sarvar Temurovich Islomov, Nodira Isroilovna Tursunova, Sitora Rakhmatova, Fatima Shokir Qizi Shodieva, Mirzarakhim Mirzakhakimovich Baynazarov, Khasanjon Kobiljon Ugli Odilov, Dilbar Yuldashevna Yuldasheva, Golib Abdulloyevich Khakimov Published: 2026 ABSTRACT AbstractBackgroundCervical cancer remains a leading cause of morbidity and mortality among women in Uzbekistan, exceeding WHO elimination targets. Despite a school-based HPV vaccination program launched in 2019, screening coverage remains low, leading to delayed diagnoses; we examined cervical cancer trends in Tashkent (2014-2024) to evaluate the impact of the Korea Foundation for International Healthcare (KOFIH) screening project and UNFPA/UNDP HPV testing initiatives.MethodsWe analyzed Tashkent City Oncoregister data for women diagnosed with invasive cervical cancer or carcinoma in situ (CIS) between 2014 and 2024 (n = 2140) after data cleaning. We calculated age-standardized incidence and mortality rates, assessed survival using Kaplan-Meier methods, and identified mortality predictors using Cox regression models.ResultsThe age-standardized incidence of invasive cervical cancer was 14.9 per 100 000. After KOFIH and UNFPA/UNDP implementation, CIS diagnoses increased 3.4-fold (0.2-4.7 per 100 000), while the proportion of advanced-stage (III-IV) disease declined by 43%. Five-year survival showed clear stage dependence, ranging from 91.5% for CIS to 27.8% for stage IV. Age and stage were independent predictors of mortality. Crude mortality declined from 10.2 to 1.1 per 100 000 from 2014 to 2024, and case fatality decreased from 48.6% (2014-2019) to 27.2% (2020-2024).ConclusionOrganized screening through KOFIH and UNFPA initiatives can improve early detection and reduce mortality. However, 76.6% of staged invasive cases remain advanced (II-IV), and more than 80% of women have never been screened, underscoring the need to expand screening, sustain HPV vaccination, and strengthen prevention programs in Central Asia. Introduction Cervical cancer persists as a major global health burden, ranking fourth in incidence among women worldwide and demonstrating stark disparities between high-income and low-middle-income nations.1 In 2020, global estimates revealed 662 301 new cases and 348 874 deaths, with mortality rates in Central Asia nearly doubling the world average.2 Uzbekistan exemplifies this regional challenge, where cervical cancer stands as the second most frequent cancer overall and the second leading cause of cancer mortality in women aged 15-44. The capital Tashkent reported an age-standardized incidence rate of 11.4/100 000 (2019 data) compared to the national rate of 14.9/100 000 in 2023, though both figures significantly exceed the World Health Organization’s (WHO) elimination target of <4/100 000.3-6 Uzbekistan has made substantial progress in primary prevention of cervical cancer while secondary prevention remains limited. The country’s school-based HPV vaccination program, launched in 2019 with WHO/UNICEF support, achieved 94% coverage among girls aged 12-14 years. This success was driven by a culturally tailored communication strategy framed as “future motherhood preservation,” which emphasized protection of girls’ future reproductive health and family life, combined with intensive school-based outreach and engagement of parents and teachers. These efforts generated strong community acceptance and placed Uzbekistan among the leading countries in Eastern Europe and Central Asia in HPV vaccination coverage.5,7 However, secondary prevention remains severely inadequate. In 2019, estimated cervical cancer screening coverage in Uzbekistan remained very low, with only 12% of women aged 25-65 and 13% of women aged 30-49 ever screened, and just 9%-11% screened in the last 3-5 years, far below the WHO target of at least 70% coverage by 2030. At present, Uzbekistan does not yet operate a fully implemented nationwide, population-based cervical screening program; instead, screening is delivered predominantly on an opportunistic basis through primary health care facilities and gynecology departments in polyclinics and hospitals, including the Republican Specialized Scientific and Practical Centre for Oncology and Radiology and its regional branches, where visual inspection with acetic acid is provided free of charge while Papanicolaou cytology, colposcopy, and HPV testing are largely available in larger clinics on a fee-for-service basis (except for certain vulnerable groups), resulting in fragmented services, low participation, and the absence of systematic follow-up for most women.5,8-10 Survival rates for cervical cancer are closely linked to the stage at diagnosis, with 5-year relative survival exceeding 90% for localized disease but dropping below 20% for cases with distant metastases.5,6 However, survival estimates can vary widely across regions and healthcare settings, reflecting differences in access to timely diagnosis, treatment modalities, and follow-up care. In recent years, survival analysis has become an essential tool for evaluating the effectiveness of cancer control programs and guiding resource allocation, particularly in settings where population-based data are limited. Tashkent’s epidemiological profile reveals important characteristics of cervical cancer burden and etiology. A 2009-2016 analysis documented 1190 cases with a mean age of 54.5 years, and mortality was particularly elevated in older women, reaching 114.2 per 100 000 in those aged ≥65 years. Recent virological studies show an 8.5% high-risk HPV prevalence nationally, with HPV16/18 accounting for 38.9% of infections. Currently, 68.9% of cervical cancers in Uzbekistan are attributable to HPV16/18 and other genotypes covered by licensed prophylactic HPV vaccines, indicating that expanded vaccination programs can substantially reduce future incidence from vaccine-targeted types. Nevertheless, a considerable proportion of cervical cancer cases in Uzbekistan at present arise from infections acquired before the introduction of HPV vaccination and from high-risk HPV genotypes for which vaccine effectiveness may be lower, so the short-term impact of vaccination on overall incidence is expected to be modest.6,11,12 Despite the global focus on cervical cancer elimination, comprehensive population-based studies examining trends in incidence, stage distribution, and survival in Central Asian contexts remain limited. Across the 5 post-Soviet Central Asian republics, a pooled 2018 analysis estimated a combined cervical cancer incidence of 12.5 and mortality of 12.3 per 100 000 women, with marked variation between countries: incidence and mortality were 5.7 and 3.2 per 100 000 in Tajikistan, 13.6 and 8.8 per 100 000 in Turkmenistan, 9.9 and 5.4 per 100 000 in Uzbekistan, and 19.9 and 10.9 per 100 000 in Kyrgyzstan.13 A recent study from the Kyrgyz Republic reported an increase in cervical cancer prevalence from 97.5 to 105.3 per 100 000 women between 2012 and 2017, with mortality among diagnosed patients ranging from 8.3% to 9.7% and pronounced regional disparities between oblasts (province-level administrative regions).14 By updating and extending registry based analyses from Tashkent, our study complements emerging evidence from Kazakhstan and Kyrgyzstan and contributes much needed population-based data on cervical cancer stage distribution and survival in Central Asia. Nevertheless, across Central Asia cervical cancer screening remains largely opportunistic, with only Kazakhstan operating a structured cytology-based program and all 5 countries showing incidence and mortality levels several-fold higher than those observed in many Western European settings, underscoring the absence of robust population-based data on stage at diagnosis and survival. This study aimed to update and expand prior analyses of cervical cancer in Tashkent by using the same population-based cancer registry to ensure comparability, while extending observation to 2014-2024. Specifically, we described temporal trends in invasive cervical cancer incidence, stage distribution, and survival, and to contrast the 2014-2019 period with 2020-2024, when United Nations Population Fund (UNFPA)- and Korea Foundation for International Healthcare (KOFIH)-supported HPV-based screening and free campaigns were introduced.11,15 Methods Study design and setting This retrospective, population-based cohort study was conducted using data from the Tashkent City Oncoregister, which systematically collects information on all newly diagnosed cancer cases among residents of Tashkent, Uzbekistan. Between 2014 and 2024, the female population aged 20 years and older increased from approximately 0.85-1.06 million, reflecting combined effects of natural population growth and net in migration to Tashkent City. The registry does not capture separate indicators of fertility and migration, so we treat these demographic changes as background context rather than as direct covariates in the incidence models. Study population and data cleaning The study included all female residents of Tashkent City diagnosed with cervical cancer or carcinoma in situ (CIS) (ICD-10 code C53) between January 1, 2014 and December 31, 2024. The raw cancer registry extract initially contained 2156 records. Four duplicate patient IDs were identified and removed, retaining the first occurrence in each case and reducing the dataset to 2152 records. An additional 12 records with impossible or implausible age information (birth year equal to 0, age >100 years, or age younger than 20 years) were then excluded, yielding a final cleaned analytic cohort of 2140 records. After data cleaning, 62 records originally coded as Stage Unknown (TxNxMx) were reclassified as having unknown stage and were excluded from all stage-specific analyses, including Kaplan-Meier survival estimates and Cox regression models. No additional statistical imputation was performed for other missing variables; analyses were restricted to records with complete data for the relevant endpoints. Survival time adjustment Survival time was defined as the interval from date of diagnosis to death or last known follow-up and was truncated at a maximum of 120 months. For patients with recorded survival time of 0 months and a lethal outcome, survival time was imputed to 0.5 months (approximately 15 days) to avoid computational artifacts in the survival models while preserving these immediate post-diagnosis deaths in the analysis. Statistical analysis Annual crude and age-specific incidence rates per 100 000 women aged 20 years and older were calculated for invasive cervical cancer and for total cervical neoplasia (invasive cancer plus CIS). Age-standardized incidence rates were derived by the direct method using the WHO World Standard Population (2000-2025). Mortality analyses were restricted to invasive cervical cancer. Because follow-up for cases diagnosed in 2024 was limited to 10 months or less, the 2024 case fatality rate is not directly comparable to estimates from earlier years. Overall survival at 1, 3, and 5 years was estimated using Kaplan-Meier methods stratified by FIGO stage, excluding cases with unknown stage. Relative survival was calculated as the ratio of observed Kaplan-Meier survival to expected survival derived from age-specific all-cause mortality rates for women in Tashkent. Cox proportional hazards models were used to identify independent predictors of mortality, adjusting for age at diagnosis and stage, and the proportional hazards assumption was assessed using Schoenfeld residuals. All statistical analyses were conducted using R, version 4.3.1, and 2-sided P-values <.05 were considered statistically significant. Results Between 2014 and 2024, we identified 2140 cervical cancer cases in Tashkent as the female population aged ≥20 years increased from 0.85 to 1.06 million. Mean age at diagnosis was 54.4 ± 12.1 years. The overall crude incidence of invasive cervical cancer over this period was 18.3 per 100 000 women (95% CI: 17.5-19.1), with an age-standardized incidence of 14.9 per 100 000 (95% CI: 14.2-15.6) (Table 1). Table 1.Annual crude and age-standardized incidence rates of cervical cancer in Tashkent, 2014-2024 (per 100 000 women, with 95% CIs).YearTotal casesInvasive casesCrude IR invasive (95% CI)ASR invasive (95% CI) 2014 14313616.0 (13.3-18.6)13.3 (11.0-15.5) 2015 15314416.7 (14.0-19.5)13.8 (11.5-16.0) 2016 16015217.5 (14.7-20.3)14.3 (12.0-16.5) 2017 17515117.2 (14.4-19.9)14.0 (11.7-16.2) 2018 20017419.5 (16.6-22.4)15.7 (13.4-18.1) 2019 19217519.3 (16.5-22.2)15.6 (13.3-17.9) 2020 22319821.5 (18.5-24.5)17.1 (14.7-19.4) 2021 23419924.4 (21.3-27.6)16.5 (14.2-18.8) 2022 25122622.4 (19.5-25.3)17.8 (15.5-20.1) 2023 22519919.2 (16.5-21.9)15.2 (13.1-17.4) 2024 18413112.4 (10.3-14.5)9.7 (8.1-11.4) 2014-2024 2140188518.4 (17.6-19.2)14.9 (14.2-15.6)Annual number of invasive cervical cancer cases and corresponding crude and age-standardized incidence rates (WHO World Standard Population, 2000-2025) per 100 000 women in Tashkent City from 2014 to 2024.Abbreviations: ASR = age-standardized incidence rates; CI = confidence interval; IR = incidence rates. Annual crude and age-standardized incidence rates of cervical cancer in Tashkent, 2014-2024 (per 100 000 women, with 95% CIs). Annual number of invasive cervical cancer cases and corresponding crude and age-standardized incidence rates (WHO World Standard Population, 2000-2025) per 100 000 women in Tashkent City from 2014 to 2024. Abbreviations: ASR = age-standardized incidence rates; CI = confidence interval; IR = incidence rates. Case numbers rose from 143 in 2014 to a peak of 251 in 2022, then declined to 184 in 2024. Crude invasive incidence rates followed a similar pattern, increasing from 16.0 per 100 000 in 2014 to 22.4 per 100 000 in 2022, before falling to 12.4 per 100 000 in 2024. When stratified by case type, invasive cancer cases increased from 136 in 2014 to 226 in 2022 and then decreased to 131 in 2024, while CIS cases rose from 0 in 2014 to 49 in 2024 (Table S1). Incidence increased steeply with age, peaking at 47.7 per 100 000 in women aged 50-64 years, while remaining consistently low in women aged 20-34 years; after 2022, incidence declined across all age groups (Figure 1). Age-specific crude incidence trends of Tashkent disease by year of diagnosis (2014-2024). At diagnosis, the stage distribution was CIS (8.9%, n = 184), stage I (14.1%, n = 291), stage II (39.1%, n = 808), stage III (30.5%, n = 632), stage IV (7.4%, n = 154), and unknown/unstaged (3.4%, n = 71; excluded from all survival analyses). Among invasive cases with known stage, 76.6% were diagnosed at advanced stage (II-IV), reflecting historically suboptimal screening coverage. Carcinoma in situ incidence increased from 0.0 per 100 000 in 2014 (no CIS cases) to 4.6 per 100 000 (95% CI: 3.4-6.1) in 2024, representing a 3.4-fold increase. This rise was modest between 2017 and 2023 (approximately 1.8-3.0 per 100 000) and then accelerated in 2024 (Table S1). Kaplan-Meier survival analysis demonstrated a clear stage-dependent gradient, with highly significant differences in survival across stages (log-rank test P < .0001). Survival probabilities at 12, 36, and 60 months declined progressively with advancing FIGO stage (Figure 2, Table S2). Survival probability of women in Tashkent with cervical cancer by cancer stage over time. Abbreviation: CIS = Carcinoma in situ. Five-year Kaplan-Meier survival rates demonstrated the critical importance of early detection: CIS 91.5% (95% CI: 85.3%-98.2%), stage I 82.2% (95% CI: 77.2%-87.6%), stage II 69.7% (95% CI: 66.1%-73.5%), stage III 43.7% (95% CI: 39.5%-48.4%), and stage IV 27.8% (95% CI: 20.7%-37.5%) (Table S2). Early-stage disease (CIS and stage I) maintained 5-year survival above 75%, whereas patients with stage IV disease had median survival below 25 months and survival probabilities dropped below 50% within the first 2 years after diagnosis (Figure 2). To account for background mortality, we additionally estimated relative survival using age- and calendar-year–specific all-cause mortality rates for women in Tashkent (Table 2). Five-year Kaplan-Meier observed survival was 91.5% for CIS, 82.2% for stage I, 69.7% for stage II, 43.7% for stage III, and 27.8% for stage IV, while corresponding 5-year relative survival estimates were 94.5%, 84.8%, 71.9%, 45.4%, and 29.0%, respectively. Relative survival thus showed a marked stage-dependent decline: stage I patients achieved 84.8% of the survival expected for age-matched women in the general population, falling to 29.0% for stage IV patients, confirming the profound impact of advanced stage at diagnosis on long-term outcomes (Table 2). Table 2.Stage-specific 1-, 3-, and 5-year observed, expected and relative survival among women with cervical cancer in Tashkent.StageNObserved 1-year survivalExpected 1-year survivalObserved 3-year survivalExpected 3-year survivalObserved 5-year survivalExpected 5-year survivalRelative 1-year survivalRelative 3-year survivalRelative 5-year survival Stage 0 (CIS) a 1840.980.990.970.980.920.970.980.990.94 Stage I 2910.930.990.870.980.820.970.940.900.85 Stage II 8080.910.990.750.980.700.970.920.760.72 Stage III 6320.750.990.500.980.440.960.760.520.45 Stage IV 1540.450.990.320.970.280.960.450.330.29aAbbreviations: CIS = Carcinoma in situ. Stage-specific 1-, 3-, and 5-year observed, expected and relative survival among women with cervical cancer in Tashkent. Abbreviations: CIS = Carcinoma in situ. Cox regression identified age at diagnosis and stage as independent predictors of mortality. Each additional year of age increased the hazard of death by 1.9% (hazard ratio [HR] 1.019, 95% CI: 1.012-1.026, P < .001), while each 1-stage increase in FIGO stage (treated as a numeric variable) was associated with a 2.10-fold higher risk of death (HR 2.104, 95% CI: 1.922-2.302, P < .001). In the categorical stage model, the HR for stage IV compared with CIS was 18.7 (95% CI: 9.8-35.9) (Figure S1, Table S3). Stage was the strongest prognostic factor; the proportional hazards assumption for stage was borderline (Schoenfeld test P = 1.2 × 10−7), which should be acknowledged as a limitation (Table S4). Of 2140 women in the analytic cohort, 712 deaths were recorded during follow-up. Crude mortality declined from 10.2 per 100 000 women in 2014 to 1.23 per 100 000 in 2024, and age-standardized mortality showed a similar reduction, from 8.4 to 0.92 per 100 000. Comparing the pre-intervention period (2014-2019) with the post-intervention period (2020-2024), mean crude mortality fell by 37% and age-standardized mortality by 39.1%. Case fatality improved substantially, decreasing from 64.0% in 2014 to 9.2% in 2024 (noting that the 2024 estimate is based on ≤10 months of follow-up and is not directly comparable with earlier years); across periods, case fatality declined from 48.6% in 2014-2019 to 27.2% in 2020-2024, a 21.4-percentage-point reduction (Figure 3). Trends in cervical cancer mortality and case fatality among women aged ≥20 years in Tashkent, 2014-2024. Annual age-standardized and crude cervical cancer mortality rates per 100 000 women aged ≥20 years (lines, left axis) and case fatality (%) (bars, right axis) in Tashkent from 2014 to 2024, with the vertical dashed line indicating the year when organized cervical cancer screening was initiated (2020). Discussion This study provides vital insights into the epidemiology, stage distribution, and survival outcomes of cervical cancer in Tashkent over an 11-year period. Our analysis shows a period-averaged age-standardized invasive cervical cancer incidence of approximately 14.9 per 100 000 women in 2014-2024, with annual ASR values ranging from 13.3 to 17.8 per 100 000 between 2014 and 2023 before dropping to 9.7 per 100 000 in 2024. Compared with the previously reported Tashkent ASR of 11.4 per 100 000 for 2009 to 2016, this represents an increase of about one-third rather than the nearly 50% originally reported, although the direction of change is unchanged. As before, this apparent rise is most plausibly explained by improved case detection through enhanced cancer registry completeness and emerging screening activities rather than a true increase in underlying disease burden.6,11,12,16 This interpretation is supported by several converging lines of evidence: the marked rise in CIS incidence from 0.35 per 100 000 in 2015 to 4.63 per 100 000 in 2024 (around a 13-fold increase), together with a decline in age-standardized invasive incidence from a peak of 17.82 per 100 000 in 2022 to 9.74 per 100 000 in 2024. This temporal pattern—substantial growth in pre-invasive detection followed by falling invasive incidence—is characteristic of successful screening program implementation.13,17-21 Tashkent’s invasive cervical cancer incidence remains higher than national estimates. The period-averaged ASR of 14.9 per 100 000 in our data exceeds the national Uzbekistan ASR of 11.3 per 100 000 reported around 2020, consistent with an urban-rural gradient in cancer detection and access to diagnostic services.7,22 Tashkent’s comprehensive oncoregistry system and centralized oncology services likely enable more complete case ascertainment than rural regions, which often lack advanced diagnostic facilities and systematic surveillance. This pattern aligns with broader Central Asian epidemiology, where cervical cancer remains one of the most common female malignancies in Kazakhstan, Kyrgyzstan, and Uzbekistan. The persistent regional burden has been attributed to fragmented screening infrastructure, delayed or incomplete HPV vaccination implementation, and suboptimal coordination of prevention and treatment services across national boundaries.1,4,13,15,17,22-24 Against this backdrop, the mortality improvements observed in Tashkent are notable. Crude cervical cancer mortality fell from 10.20 per 100 000 in 2014 to 1.13 per 100 000 in 2024, corresponding to an approximate 89% reduction over the study period. Period comparison shows that mean crude mortality decreased from 8.63 per 100 000 in 2014 to 2019 to 5.34 per 100 000 in 2020 to 2024, a relative reduction of about 38%. Although age-standardized mortality estimates are not directly tabulated in the shared spreadsheet, the previously calculated decline from 8.40 to 0.92 per 100 000 and a 39.1% reduction between 2014-2019 and 2020-2024 (6.95-4.24 per 100 000) remains directionally consistent with the crude rates. These values are substantially lower than reported age-standardized cervical cancer mortality rates in Kazakhstan and Kyrgyzstan (approximately 9-10 per 100 000), suggesting that systematic implementation of cervical cancer screening projects since 2020—together with Tashkent’s relatively stronger healthcare infrastructure—can meaningfully reduce cervical cancer deaths even in Central Asia’s challenging epidemiological context.11,13,14,16,21,25-27 The observed fluctuations in incidence over time warrant careful interpretation. Age-standardized invasive incidence rose from 13.28 per 100 000 in 2014 to a peak of 17.82 per 100 000 in 2022, before declining sharply to 9.74 per 100 000 in 2024. Short-term fluctuations in crude and age-standardized mortality around 2016-2018 should be interpreted cautiously, as they occur in the context of relatively small annual case numbers and progressive improvements in linkage between the Tashkent City Oncoregister and vital statistics data. These variations are therefore more compatible with stochastic year to year variation and evolving data completeness than with abrupt underlying changes in disease risk. A similar pattern is seen in the crude invasive incidence rate, which increased to 22.38 per 100 000 in 2022 and then fell to 12.37 per 100 000 in 2024. This 3-phase pattern—initial incidence surge, plateau, and subsequent decline—likely reflects a combination of improved detection through expanded screening capacity (particularly in 2020-2022), depletion of a pool of previously undiagnosed prevalent cases, and emerging true incidence reduction through early detection and treatment in 2023-2024.13,18-21,23,25,27 The peak years coincide with the KOFIH screening project and UNFPA/UNDP HPV testing initiatives, while the approximately 45% reduction in age-standardized incidence between 2022 and 2024 suggests both the resolution of screening-related prevalence effects and potential contributions from pandemic-related diagnostic disruptions.11,15,26 The COVID 19 pandemic, beginning in 2020, is likely to have contributed to temporary diagnostic and treatment disruptions, consistent with international evidence showing reduced screening attendance, delayed diagnoses, and short-term changes in cancer incidence and mortality during and immediately after service interruptions. In Tashkent, the peak in invasive incidence and stage III-IV disease in 2020-2022 coincides with both the scale up of screening projects and pandemic related healthcare strain, while the subsequent decline in incidence and mortality in 2023-2024 may reflect a combination of depletion of previously undiagnosed prevalent cases, partial restoration of screening and treatment pathways, and ongoing registry strengthening. Similar patterns of temporary incidence increases followed by declines have been documented internationally when screening and prevention programs are scaled up after periods of limited coverage.18-21,26-28 The stage distribution of cervical cancer cases in our cohort, with stage II (37.8%) and stage III (29.5%) predominating among all registered cases, reflects the delayed diagnosis profile typical of settings with suboptimal screening coverage. Across the entire period, CIS accounted for 8.6% and stage I for 13.6% of all cases, while stage IV constituted 7.2% and cases with unknown stage 3.3%. When restricted to cases with known stage, stages II-IV together represented approximately 77% of diagnoses, confirming that most women still present with invasive disease beyond stage I. Over time, CIS incidence was low but steadily rising from 2015 onward, fluctuating between about 1.5 and 2.6 per 100 000 from 2017 to 2023 and then increasing more sharply to 4.63 per 100 000 in 2024, which is consistent with recent improvements in diagnostic sensitivity or reporting but does not yet demonstrate a long, smooth upward trend. In contrast, the combined crude incidence of stages III-IV disease rose to a maximum of 11.38 per 100 000 in 2020 and then declined to 4.72 per 100 000 in 2024, with the mean combined stage III-IV rate falling from approximately 7.24 per 100 000 in 2014-2019 to 6.17 per 100 000 in 2023-2024 (about a 15% reduction), indicating moderate but meaningful reduction in advanced-stage burden rather than the larger decline originally estimated. This divergent temporal pattern—rising pre-invasive detection with modest but real decreases in advanced-stage incidences compatible with the expected epidemiological signature of an emerging but still incomplete screening program. This stage migration is further clarified when examining stage distribution by period. Between 2014-2019 and 2020-2024, the proportion of CIS among known-stage cases increased from 5.3% to 12.2%, whereas stage I remained relatively similar (15.5% vs 12.7%) and stages II and III proportions decreased only slightly (40.4%-37.8% for stage II; 31.9%-29.3% for stage III), with stage IV increasing from 6.8% to 8.0%. Thus, the main stage shift is currently driven by increasing detection of CIS rather than a dramatic redistribution from stages III to IV to earlier stages, and a substantial proportion of women still present with locally advanced or metastatic disease. Nonetheless, in combination with the strong mortality and case-fatality improvements, these patterns support a genuine improvement in early detection rather than simple reclassification or artifacts of staging practice, extending previous findings from Tashkent that identified cervical cancer as the second most common female cancer.6,23,29-34 The survival analysis demonstrates a pronounced stage-dependent gradient that mechanistically explains how even modest shifts toward earlier stages can yield substantial population-level gains. Five-year overall survival probabilities declined from 91.5% for CIS to 82.2% for stage I, 69.7% for stage II, 43.7% for stage III, and 27.8% for stage IV, closely mirroring international experience with similar FIGO stage groupings.1,29-33 stage IV patients had a median survival below 25 months, whereas stage I patients consistently maintained 5-year survival above 75%, underscoring the critical importance of early detection.1,25-28 These findings are concordant with recent studies showing markedly higher survival in women treated at stages IIB-IIIA compared with those diagnosed at stages IVA-IVB, and the increasing proportion of early-stage and in situ disease in our later period provides a plausible explanation for the observed mortality declines.35-37 Therefore, it is important to note that our estimates for CIS represent overall and relative survival, based on all-cause mortality, rather than cervical cancer-specific survival. In the absence of detailed treatment and cause-of-death data, some deaths among women classified with CIS likely reflect competing comorbid conditions or misclassification rather than failure of local excisional therapy. Consequently, the 5-year relative survival of 94.5% for CIS is broadly consistent with the near-optimal outcomes reported in organized screening programs, while still acknowledging data limitations in our setting. Age emerged as an independent determinant of survival in our Cox models. Women aged 20-29 years had near-optimal outcomes with minimal decline in 5-year survival, whereas survival worsened progressively with advancing age, falling below 50% within 4 years for women aged ≥65 years. The multivariable Cox regression confirmed these relationships, with age at diagnosis associated with a 1.9% increase in mortality risk per additional year (HR 1.0189) and each incremental increase in stage (treated ordinally) associated with more than a 2-fold increase in mortality risk (HR 2.10). These age-related disparities mirror patterns seen in large registry-based analyses, where older women are more likely to present with advanced disease, have higher comorbidity burdens, reduced treatment tolerance, and face delays in detection and management.33-38 The dominance of stage as a mortality predictor in our data reinforces why even partial stage shifts toward earlier detection can have disproportionate impact on population health outcomes.31,32 The most compelling evidence of screening effectiveness in this cohort comes from the combination of sharp mortality decline and substantial improvement in case fatality despite initially rising incidence. Case fatality decreased from 64.0% in 2014 to 9.2% in 2024, with period-specific case fatality falling from 48.6% in 2014 to 2019 to 27.2% in 2020 to 2024. In the pre-screening era, nearly 1 in 2 diagnosed women died from cervical cancer; in the later period, this proportion fell to roughly 1one in 4. International evidence strongly supports the link between organized screening and reductions in both incidence and mortality—estimates from England, for example, suggest that regular screening prevents about 70% of cervical cancer deaths, with the potential to avert up to 83% under optimal participation. The striking feature of our data is that mortality fell even as incidence initially increased, a pattern most consistent with improved detection and treatment of existing disease rather than a simple decline in disease occurrence alone.13,18,19,27 Organized cervical cancer screening in Uzbekistan is still in an early and uneven stage of development. Pilot HPV-based screening projects and free screening campaigns have been introduced in selected regions, including Tashkent, since 2021-2024, yet national survey data indicate that more than 80% of women report never having been screened. Our findings are consistent with this picture: the increase in CIS diagnoses likely reflects contributions from recent targeted screening activities in our catchment area, but the persistently high proportion of stages II-IV disease and low lifetime screening uptake both indicate that overall coverage and effectiveness remain suboptimal.11,12,16 Similar challenges are evident across Central Asia, where only Belarus and Turkmenistan have reportedly achieved the WHO target of at least 70% screening coverage; in most other countries, including Uzbekistan, coverage estimates range widely from 2% to 66%.15,17,23 Cervical cancer remains a major public health problem in Eastern Europe and Central Asia, where it is the second most deadly cancer among women of reproductive age and is responsible for more than 32 000 new diagnoses and 16 000 deaths annually. The Regional Alliance for Cervical Cancer Prevention in Eastern Europe and Central Asia, launched by UNFPA in 2021, is a critical platform for coordinated action on HPV vaccination, screening, and treatment, and our Tashkent results demonstrate that meaningful mortality reductions are achievable even before full regional scale-up. However, the high residual burden of advanced disease in our cohort underscores that further investments are needed in primary prevention (HPV vaccination), screening program expansion, quality assurance, and timely access to guideline-concordant treatment.4,11,12,15,16,23,39 Several limitations qualify for interpretation of our findings. First, our analysis is geographically restricted to Tashkent, which has better-resourced oncology services, diagnostics, and cancer registry systems than other regions of Uzbekistan; consequently, these results may overestimate screening impact and underestimate true national incidence and mortality. Second, the retrospective design, based on clinical records and routine follow-up, may have incomplete vital status ascertainment, though linkage to the Tashkent City Oncoregister likely mitigates this risk. Third, our dataset lacks detailed individual-level information on screening history, HPV genotyping, specific treatment modalities, adherence to therapy, and socioeconomic variables, all of which can influence both stage at diagnosis and survival and could help explain any remaining heterogeneity in outcomes. Fourth, although 11 years of follow-up allow assessment of medium-term trends, this horizon remains relatively short for establishing whether observed improvements will be sustained, especially as HPV-vaccinated cohorts begin to reach screening age. Finally, most women in our cohort were diagnosed before the widespread rollout of HPV vaccination, precluding direct assessment of vaccination effects on incidence; the full impact of HPV vaccination will likely only become apparent in Tashkent from approximately 2034 onward. Because our analysis is based on population-level registry data without individual-level information on screening histories or treatment, we cannot attribute the observed mortality and case-fatality reductions exclusively to screening. The temporal association between the expansion of HPV-based screening projects, increased detection of CIS, moderate shifts in stage distribution, and declining case fatality is consistent with screening effects, but other factors—including improved access to evidence-based treatment, better follow-up of patients, and enhanced data completeness—almost certainly contribute. Future research priorities include deterministic linkage with national death registries to improve completeness of mortality follow-up and enable more refined period- and cohort-based survival analyses. Future analyses could also apply joinpoint regression to registry data to formally identify changes in incidence and mortality trend slopes, helping distinguish screening-related prevalence effects from longer-term secular shifts and strengthening causal inference regarding prevention interventions. Studies incorporating HPV genotype distribution, screening histories, socioeconomic indicators, and detailed treatment data are needed to identify modifiable determinants of late-stage diagnosis—particularly in older and underscreened women—and to guide tailored interventions. Prospective monitoring of cervical cancer incidence in cohorts offered HPV vaccination, beginning in the mid-2030s, will be essential to quantify long-term vaccine impact in Uzbekistan’s context. Finally, comparative analyses across Central Asian countries—leveraging emerging initiatives such as the Regional Alliance for Cervical Cancer Prevention—could help identify best practices, clarify the relative contributions of vaccination vs screening, and inform rational, equity-oriented resource allocation strategies in the region. Conclusion This 11-year, population-based study from Tashkent shows that organized cervical cancer screening can substantially improve outcomes in Central Asia even when incidence remains relatively high (average age-standardized invasive incidence ≈14.9 per 100 000). Following screening implementation in 2020, crude mortality declined from 10.2 to 1.1 per 100 000 and mean case fatality fell from 48.6% to 27.2%, despite roughly 77% of staged invasive cases still presenting at stages II-IV and >80% of women in Uzbekistan having never been screened. Scaling up high-quality HPV-based screening, ensuring broad HPV vaccination coverage, strengthening rural services, and guaranteeing equitable access to timely treatment are essential to move from Tashkent’s encouraging progress toward national and regional cervical cancer elimination. Supplementary Material