PREDICT-GTN 2: Two-factor streamlined models match FIGO performance in gestational trophoblastic neoplasia.

OBJECTIVE: The International Federation of Gynecology and Obstetrics (FIGO) scoring system uses the sum of eight risk-factors to predict single-agent chemotherapy resistance in Gestational Trophoblastic Neoplasia (GTN). To improve ease of use, this study aimed to generate: (i) streamlined models that match FIGO performance and; (ii) visual-decision aids (nomograms) for guiding management. METHODS: Using training (n = 4191) and validation datasets (n = 144) of GTN patients from two UK specialist centres, logistic regression analysis generated two-factor models for cross-validation and exploration. Performance was assessed using true and false positive rate, positive and negative predictive values, Bland-Altman calibration plots, receiver operating characteristic (ROC) curves, decision-curve analysis (DCA) and contingency tables. Nomograms were developed from estimated model parameters and performance cross-checked upon the training and validation dataset. RESULTS: Three streamlined, two-factor models were selected for analysis: (i) M1, pre-treatment hCG + history of failed chemotherapy; (ii) M2, pre-treatment hCG + site of metastases and; (iii) M3, pre-treatment hCG + number of metastases. Using both training and validation datasets, these models showed no evidence of significant discordance from FIGO (McNemar's test p > 0.78) or across a range of performance parameters. This behaviour was maintained when applying algorithms simulating the logic of the nomograms. CONCLUSIONS: Our streamlined models could be used to assess GTN patients and replace FIGO, statistically matching performance. Given the importance of imaging parameters in guiding treatment, M2 and M3 are favoured for ongoing validation. In resource-poor countries, where access to specialist centres is problematic, M1 could be pragmatically implemented. Further prospective validation on a larger cohort is recommended.

High HIV prevalence in gestational trophoblastic neoplasia patients.

OBJECTIVE: To assess the HIV prevalence in patients diagnosed with gestational trophoblastic neoplasia (GTN). DESIGN: A retrospective single centre cohort study. METHODS: A database from the Sheffield Trophoblastic Disease Centre (STDC), Sheffield, UK was searched between 1st January 2015 and 31st December 2021. A total of 3,591 patients were referred to STDC with a diagnosis of gestational trophoblastic disease (GTD), of which 221 (6.2%) were treated for GTN. The prevalence of HIV-positive tests in GTN patients was assessed. RESULTS: HIV testing was performed in 93% GTN patients ( n  = 205/221). Overall, 1.3% of GTN patients ( n  = 3/221) were HIV-positive, involving two known HIV-positive patients and one new diagnosis. This equates to a HIV prevalence of 14 : 1000, which is ∼7 to 9× higher than the HIV prevalence in Sheffield (1.9 per 1000) and Yorkshire and Humber (1.5 per 1000). CONCLUSION: Given the extremely high HIV prevalence in our population, 'opt out' HIV testing is recommended within our specialist trophoblastic centre for all referred GTD and GTN patients. There is little reason to suspect that the prevalence of HIV-positive patients is any lower in the cohort of GTD patients referred to specialist trophoblastic centres for hCG screening alone, without requiring chemotherapy, particularly considering that most GTN arises from GTD.

Factors associated with failed 'test of cure' in the NHS Cervical Screening Programme: A retrospective cohort study.

OBJECTIVE: To determine predictive factors associated with failed 'test of cure' (TOC) in the NHS Cervical Screening Programme (NHSCSP). METHODS: Retrospective cohort study of all patients treated by large loop excision of transformation zone (LLETZ) between 1st April 2014 and 1st April 2019. Those with no documented HPV genotype on referral, no TOC outcome, those having a hysterectomy, chemotherapy and/or radiotherapy were excluded from final analysis. RESULTS: Patients referred with a singular HPV genotype of HPV 16, HPV 18, or HPV Other types (HPV O) were significantly more likely to pass TOC than those referred with multiple HPV genotypes (p < 0.0001). Those with HPV genotypes including HPV O were significantly more likely to fail TOC as compared to those with genotypes of solely HPV 16 and/or 18 (p < 0.0001). Patients aged ≥51 years were significantly more likely to fail TOC when compared to all other age groups (p < 0.0001). CONCLUSION: Age >51 yrs. and infection with multiple hr-HPV types were predictors of post treatment hr-HPV persistence. Knowledge of HPV genotype both at referral, and following treatment, could allow a more individualised, and patient-centred, approach to both the management and follow up of CIN. HPV genotype should be reported as standard on all cervical screening sample results. The term HPV O should not be utilised and instead actual HPV genotype should be reported. This would enable us to optimise not only future research but would also allow future monitoring of the efficacy of vaccination programmes.

Flat-dose versus weight or body surface area-based methotrexate dosing in low-risk gestational trophoblastic neoplasia.

BACKGROUND: Single-agent methotrexate (MTX) is commonly used as first-line treatment for low-risk gestational trophoblastic neoplasia (LR-GTN), although no international consensus exists on the optimal treatment regimen to maximise complete hCG response (CR) and minimise relapse rates. Current regimens differ in the route of administration, dose scheduling, and use of flat-dose, body surface area (BSA)- or weight-based dosing. In the UK a methotrexate-folinic acid (MTX-FA) 8-day 50 mg intramuscular flat-dose regimen is used, with 15 mg oral folinic acid rescue. In LR-GTN patients, we aim to determine the effect of MTX dose adjustment by BSA and weight upon chemotherapy response and disease relapse. METHODS: Between January 1973 and August 2020, 935 LR-GTN patients treated with first-line MTX-FA were identified from a single UK specialist trophoblastic centre. Of these, 364 were included, of which 178 (49%) had a CR to first-line MTX-FA. Subgroup analyses were performed upon: (i) patients who changed chemotherapy due to MTX toxicity (n = 33); and (ii) patients with a FIGO score of 5-6 (n = 85). Logistic regression analysis explored the relationship between BSA or weight adjusted MTX dosing and: (i) CR to first-line chemotherapy; (ii) incidence of disease relapse. Linear regression analyses assessed the correlation of BSA and weight with the number of MTX-FA cycles required to achieve CR. RESULTS: In LR-GTN patients, BSA and weight adjusted MTX-FA dosing did not influence CR to first-line chemotherapy or the incidence of disease relapse. The number of MTX cycles required to achieve CR was not associated with BSA or weight. These findings were maintained in a subgroup analysis of FIGO 5-6 patients. The incidence of MTX toxicity was not influenced by BSA or weight. CONCLUSIONS: In the treatment of LR-GTN, dose individualisation using BSA or weight is not required, and fixed dosing continues to be preferred as the UK standard.

PREDICT-GTN 1: Can we improve the FIGO scoring system in gestational trophoblastic neoplasia?

Gestational trophoblastic neoplasia (GTN) patients are treated according to the eight-variable International Federation of Gynaecology and Obstetrics (FIGO) scoring system, that aims to predict first-line single-agent chemotherapy resistance. FIGO is imperfect with one-third of low-risk patients developing disease resistance to first-line single-agent chemotherapy. We aimed to generate simplified models that improve upon FIGO. Logistic regression (LR) and multilayer perceptron (MLP) modelling (n = 4191) generated six models (M1-6). M1, all eight FIGO variables (scored data); M2, all eight FIGO variables (scored and raw data); M3, nonimaging variables (scored data); M4, nonimaging variables (scored and raw data); M5, imaging variables (scored data); and M6, pretreatment hCG (raw data) + imaging variables (scored data). Performance was compared to FIGO using true and false positive rates, positive and negative predictive values, diagnostic odds ratio, receiver operating characteristic (ROC) curves, Bland-Altman calibration plots, decision curve analysis and contingency tables. M1-6 were calibrated and outperformed FIGO on true positive rate and positive predictive value. Using LR and MLP, M1, M2 and M4 generated small improvements to the ROC curve and decision curve analysis. M3, M5 and M6 matched FIGO or performed less well. Compared to FIGO, most (excluding LR M4 and MLP M5) had significant discordance in patient classification (McNemar's test P < .05); 55-112 undertreated, 46-206 overtreated. Statistical modelling yielded only small gains over FIGO performance, arising through recategorisation of treatment-resistant patients, with a significant proportion of under/overtreatment as the available data have been used a priori to allocate primary chemotherapy. Streamlining FIGO should now be the focus.

Clinical indications referrals - what is the risk of premalignant and malignant female genital tract disease in women referred to a dedicated nurse-led colposcopy service? A retrospective cohort study.

Post-coital bleeding (PCB) is a poor predictive factor for cancer and should not be managed as urgent referral. Urgent referral to colposcopy is justified however, in the presence of a visible suspicion of cervical cancer. This retrospective cohort study of women attending a clinical indications referral service aims to identify the risk of pre-malignant and malignant disease in women with clinical indication referrals to colposcopy. Thirty-seven of 3521 women (1%) were diagnosed with pre-malignant cervical or endometrial disease; 14 women (0.4%) were diagnosed with cancer (11 cervix, three endometrial). To detect one cancer in women referred with an abnormal cervix, one would need to see 70 women; to detect one cancer in women referred with PCB one would need to see 790 women. Improved education in primary care and obstetrics and gynaecology training is key to improving clinical indications referral services, which is otherwise an effective and efficient service.Impact StatementWhat is already known on this subject? Post-coital bleeding is a poor predictive factor for cancer and should not be considered an urgent referral.What do the results of this study add? The presence of a visible suspicion of cervical cancer however does warrant urgent referral as approximately one in 70 women will have a malignancy detected.What are the implications of these findings for clinical practice and/or further research? Improved education in primary care and obstetrics and gynaecology training is the key to improving clinical indications referral services.

Invasive cervical cancer audit: What lessons can we learn locally and where would we stand with regard to duty of candour?

OBJECTIVE: To identify lessons learned locally from the invasive cervical cancer audit. To estimate the impact that the application of 'Duty of Candour' may have upon our future service provision. METHODS: Retrospective cohort study with interval analysis of all women diagnosed with cervical cancer at Sheffield Teaching Hospitals NHS Foundation Trust between 1 April 2007 to 31 December 2019. Data were collected prospectively with retrospective categorisation by screening history and invasive cervical cancer audit outcomes as satisfactory, satisfactory with learning points, and unsatisfactory. Statistical analysis was performed using the chi-squared test and paired t-test. RESULTS: Cervical cancer was diagnosed in 344 women. Seventy-eight (23%) had no record of prior cervical cytology, 108 (31%) had delayed attendance to the screening programme, 102 (30%) were detected by routine screening, and 56 (16%) were screening programme compliant. Satisfactory management was undertaken in 301 (87.5%) cases, 26 cases (7.5%) were satisfactory with learning points, and 17 cases (5%) were considered as unsatisfactory. CONCLUSIONS: Seventeen cases were applicable to the Duty of Candour process equating to 1.3 cases per year, incurring minimal impact upon future service provision. Invasive audit categorisation is subject to bias, however, with the potential for considerable intra- and inter-observer variation; the authors accordingly recommend that a further study be conducted to investigate both the consistency and reproducibility of the invasive cervical cancer audit categorisation.

Are we managing our patients correctly following treatment for cervical glandular intraepithelial neoplasia? A review of practice at the Jessop Wing Colposcopy Unit.

OBJECTIVE: Women diagnosed with cervical glandular intraepithelial neoplasia (CGIN) remain at risk of further pre-malignant and malignant disease and require rigorous post-treatment follow-up. We assess the effectiveness and safety of community cervical sampling follow-up in women treated for CGIN. METHODS: A retrospective study was conducted of women diagnosed with CGIN between April 1, 2013, and March 31, 2019, at Jessop Wing Colposcopy Unit, Sheffield, UK. RESULTS: Of 140 women diagnosed with CGIN, 76 had co-existing cervical intraepithelial neoplasia (CIN). Cytologists were significantly more likely to report glandular neoplasia in the absence of co-existing CIN, and high-grade dyskaryosis in its presence (Ps < 0.0001). Co-existing CIN was significantly more likely to be present with high or low-grade compared to normal colposcopy findings (P < 0.0001). The 6-month test of cure (TOC) was attended by 67% of women (84% within 12 months), and the 18-month post-treatment sampling by 52.5% of women (70% within 24 months). Colposcopy recalled 96% of women correctly for the 18-month sampling, but 20% of women undertaking primary care samples were incorrectly recalled at 3 years instead. CONCLUSIONS: When CGIN is diagnosed, two dates for recall should be provided at 6 and 18 months post-treatment to the Cervical Screening Administration Service and the centralised screening laboratory ensuring the 18-month post-treatment sample is correctly appointed, preventing women with HPV-negative TOC samples being returned to 3-year recall. Follow-up of CGIN should be closely audited by the centralised laboratories ensuring women with CGIN are not put at additional risk.

Computed tomography chest imaging offers no advantage over chest X-ray in the initial assessment of gestational trophoblastic neoplasia.

BACKGROUND: The International Federation of Gynaecology and Obstetrics (FIGO) score identifies gestational trophoblastic neoplasia (GTN) patients as low- or high-risk of single-agent chemotherapy resistance (SACR). Computed tomography (CT) has greater sensitivity than chest X-ray (CXR) in detecting pulmonary metastases, but effects upon outcomes remain unclear. METHODS: Five hundred and eighty-nine patients underwent both CXR and CT during GTN assessment. Treatment decisions were CXR based. The number of metastases, risk scores, and risk category using CXR versus CT were compared. CT-derived chest assessment was evaluated as impact upon treatment decision compared to patient outcome, incidence of SACR, time-to-normal human chorionic gonadotrophin hormone (TNhCG), and primary chemotherapy resistance (PCR). RESULTS: Metastasis detection (p < 0.0001) and FIGO score (p = 0.001) were higher using CT versus CXR. CT would have increased FIGO score in 188 (31.9%), with 43 re-classified from low- to high-risk, of whom 23 (53.5%) received curative single-agent chemotherapy. SACR was higher when score (p = 0.044) or risk group (p < 0.0001) changed. Metastases on CXR (p = 0.019) but not CT (p = 0.088) lengthened TNhCG. Logistic regression analysis found no difference between CXR (area under the curve (AUC) = 0.63) versus CT (AUC = 0.64) in predicting PCR. CONCLUSIONS: CT chest would improve the prediction of SACR, but does not influence overall treatment outcome, TNhCG, or prediction of PCR. Lower radiation doses and cost mean ongoing CXR-based assessment is recommended.

Conservative management of CIN2: National Audit of British Society for Colposcopy and Cervical Pathology members' opinion.

There is no doubt that organised cervical screening programmes have significantly reduced the rates of cervical cancer by detection and treatment of high-grade cervical intraepithelial neoplasia (CIN2, CIN3). National UK guidelines do not differentiate between CIN2 and CIN3 as separate entities and recommend treatment for both, although a degree of uncertainty exists regarding the natural history of CIN2. This national survey of British Society for Colposcopy and Cervical Pathology members aimed to assess attitudes towards conservative management (CM) of CIN2 in the UK and identify potential selection criteria. In total, 511 members responded (response rate 32%); 55.6% offered CM for selective cases; 12.4% for all cases; 16.4% had formal guidelines. Most agreed age group was >40yrs (83%), HPV 16/18 positive (51.4%), smoking (60%), immuno-compromise (74.2%), and large lesion size (80.8%) were relative contraindications for CM. 75.9% favoured six-monthly monitoring, with 80.2% preferring excisional treatment for persistent high-grade disease. Many UK colposcopists manage CIN2 conservatively without formal guidelines. Potential selection criteria should be investigated by a multicentre study. Impact statement Although anecdotally some colposcopists manage many women with CIN2 conservatively, this National Audit of British Society for Colposcopy and Cytopathology members, we believe, is the first time this has been formally recorded. The survey assesses current attitudes towards conservative management (CM) of CIN2 and seeks to identify potential selection criteria that could be used to identify suitable women. It received over 500 responses and significantly, identified many colposcopists recommending CM of CIN2 for patients despite the lack of any formal guidance regarding this approach. The greater majority of respondents were keen to consider participating in a multicentre trial on CM of CIN2 targeting the UK screening population (25-64 years). The paper has international relevance as ACOG and ASCCP have recently changed their guidance for the management of CIN2 in younger women and now recommend CM with monitoring rather than first line ablative or excisional treatment due to concerns regarding overtreatment, especially in women who have not yet completed their family.

Development and Psychometric Testing of an Electronic Patient-Reported Outcome Tool for Vulval Disorders (ePAQ-Vulva).

OBJECTIVE: Development of an electronic patient-reported outcome measure (PROM) specifically designed for vulval disorders. Psychometric testing of the components of the questionnaire, which assess vulval symptoms, sexual function, and quality of life (QoL). MATERIALS AND METHOD: Development and programming of the instrument (ePAQ-Vulva) was informed by national guidelines for the assessment of vulval disorders, an expert panel, and a survey of 61 vulval clinic patients. The PROM assesses frequency and impact of vulval symptoms, sexual function, and QoL. It also records conditions and behaviors related to vulval disorders and patient concerns/goals.Scale generation and psychometric testing were undertaken for the vulval symptoms, sexual function, and QoL components of the PROM with 91 participants; descriptive statistics, factor analysis and internal reliability of identified domains, and agreement between free-text and multiple-choice items to assess convergent validity and interrater reliability of picture items were assessed. RESULTS: Descriptive statistics showed high floor effects for seven questionnaire items. Factor analysis identified 5 principal components. These were reviewed and amended to provide a putative domain structure of 6 domains. Internal reliability of these domains was assessed using Cronbach α, producing values of 0.715 to 0.917. Interrater reliability of the picture items produced a κ statistic of 0.405. Spearman rank showed moderate correlation between multiple-choice answers and free-text concerns (r = 0.364-0.462) in 3 of the 6 domains (pain, sex, and dyspareunia). CONCLUSIONS: ePAQ-Vulva offers the first patient-reported outcome tool, specifically designed for vulval disorders. The instrument requires further validation and testing, including evaluation of the stability, responsiveness, and reliability.

Influence of high risk HPV genotype on colposcopic performance: A large prospective study demonstrates improved detection of disease with ZedScan I, particularly in non-HPV 16 patients.

OBJECTIVE: To assess the influence of high-risk Human Papilloma Virus (hrHPV) genotyping on the detection of high-grade disease (CIN2+) using colposcopic impression both with and without electrical impedance spectroscopy (ZedScan I) as an adjunct. STUDY DESIGN: A prospective cohort of women with a known hrHPV genotype referred to a single colposcopy service. RESULTS: 839 women underwent colposcopy and ZedScan I examination. 613 women were referred with abnormal cytology; 411 (67%) with low-grade dyskaryosis (67%) and 202 (33%) with high-grade dyskaryosis. 187 were referred with persistent hrHPV but negative cytology. 35 were attended for follow up and 4 for a clinical indication. 159 (19%) women were positive for HPV16 only; 54 (6%) with HPV18 only, 443 (53%) women were positive for hrHPV other types (HPV O). 183 (22%) were positive for multiple hrHPV genotypes. CIN2+ was present in 170 (84.2%) of high-grade and 69 (16.7%) of low-grade cytology referrals. Colposcopy was better at detecting HPV16 associated CIN2+ than that associated with HPV18 or HPV O (86.9% vs 79.7%, p=0.0191). ZedScan I increased the detection of CIN2+ from 85.6% to 96% irrespective of hrHPV genotype status (p<0.0001). CONCLUSION: The use of an electrical impedance spectroscopic device (ZedScan I) increases detection of CIN2+ irrespective of hrHPV genotype.

Healthy women with persistently elevated hCG levels: a case series of fourteen women.

OBJECTIVE: To review our own data and that in the literature in order to assess likely morbidity and mortality risks and enhance the information that we can provide to patients suffering with this condition. STUDY DESIGN: This was a retrospective case series using data from the Sheffield Trophoblastic Disease Centre Database combined with data from prior publications. RESULTS: A diagnosis of elevated human chorionic gonadotropin (hCG) in an otherwise healthy woman carries an 11-19% risk of malignancy and 1-3% risk of mortality. Irrespective of persistently elevated hCG, however, pregnancy appears to be a viable and safe prospect. CONCLUSION: Persistently elevated hCG in healthy, nonpregnant women is a rare clinical scenario. Due to the rarity of this condition and potential future malignancy risk, we believe that reporting of future cases is crucial, as is a liaison between national and international trophoblastic disease centers, if we are to gain a more thorough understanding of this possibly premalignant condition.

Pregnancy following vulvar squamous cell carcinoma: a report of two cases.

Pregnancy following squamous cell carcinoma of the vulvar is rare. Its rarity is reflected by a paucity of cases reported in the literature. We report two cases of pregnancy following diagnosis and treatment for vulvar squamous cell carcinoma, and review eleven prior reported cases. In successfully treated vulvar cancer subsequent pregnancy is not shown to increase the risk of disease recurrence, and there appears to be no deleterious effects during the antenatal period. It is possible, when considering prior reports, that prior vulvectomy may increase the likelihood of delivery by caesarean section, though modifications in the surgical management of vulvar carcinoma may have decreased this risk.

Influence of age as a factor in the outcome of gestational trophoblastic neoplasia.

OBJECTIVE: To question whether older patients have a worse prognosis or poorer outcomes with chemotherapeutic regimens. STUDY DESIGN: All gestational trophoblastic disease (GTD) cases registered between January 1986 and September 2006 (n = 8,536) were reviewed and stratified for age. Chi2 analysis was used to ascertain whether significant differences existed with regard to patient age and histologic diagnosis or treatment requirement. Logistic regression analysis was used to predict chemotherapeutic outcomes in patients > 40 years age (n = 50). RESULTS: An increased relative risk of high-risk pathology and need for treatment in the > 40 years age-group was found. Modification of the World Health Organization risk by removing the age score or altering the age score was significant on univariate analysis but did not actually improve the predictive ability with regard to patient treatment outcomes either with first-line or overall therapy. CONCLUSION: Patient age may not be a risk factor for gestational trophoblastic neoplasia. With birth rates in women > 40 years and maternal age at first pregnancy significantly increasing in the United Kingdom, it is important to improve our understanding of the relationship between GTN and maternal age.

Epithelioid trophoblastic tumor: a review of the literature.

OBJECTIVE: To identify common characteristics and provide suggestions for future reporting and management of epithelioid trophoblastic tumors (ETTs). STUDY DESIGN: Definitions and treatment strategies are unclear because of low incidence and paucity of reported data. Literature search revealed 52 cases of ETT; 67% presented with abnormal vaginal bleeding, 36% had prior evidence of molar pregnancy and 35% presented with metastases. Mean age at diagnosis was 38 years. Mean pregnancy interval was 76 months. Human chorionic gonadotropin levels were 12-148,460 IU/L. RESULTS: Histologic and immunohistochemical reporting varied markedly between centers, as did treatment regimens. A total of 13% were reported as dead from disease, though duration of follow-up was variable (range, 1-39 months). Differentiation of prognostic factors in ETT is problematic. Most reported cases lack long-term follow-up, and disease recurrence in ETT can be late and complex. Distinguishing ETT from other diagnoses may lead to underreporting, with an adverse prognosis associated with diagnostic delay. CONCLUSION: Case reporting should contain detailed information on clinicopathologic, histologic and immunohistochemical characteristics and treatment. Data centralization in these rare tumors may be beneficial in identifying relevant prognostic parameters.

The prognostic and predictive value of syntactic structure analysis in serous carcinoma of the ovary.

The objective of this study was to determine whether syntactic structure analysis (SSA) can predict survival outcome and chemotherapeutic response in ovarian carcinoma. Syntactic structure analysis parameters, blindly determined in archived hematoxylin and eosin sections of 132, International Federation of Gynecology and Obstetrics (FIGO) stage I to IV serous ovarian tumors, and clinicopathologic parameters were evaluated as to their univariate and multivariate prognostic value and ability to predict chemotherapy response as measured by changes in CA125 levels. Univariate analysis revealed FIGO stage, tumor grade, preoperative CA125, presence of ascites, extent of disease residuum, and the SSA parameters minimum spanning tree (min MST), maximum MST (max MST), percent connectivity to 1, and 2 nearest neighbors to be significant predictors of overall survival and disease-free survival. Tumor grade, FIGO stage, extent of disease residuum, presence of ascites, and percent connectivity to 2 nearest neighbors were found to be significant predictors of chemotherapy response. Multivariate analysis revealed extent of disease residuum to be a significant predictor for overall survival (P

Prognostic value of measurements of angiogenesis in serous carcinoma of the ovary.

The study objective was to determine whether tumor vascularity correlates with patient survival, to compare newer semiautomated methods of angiogenesis assessment to older methods, and to determine if advanced image analysis methods can offer useful patient outcome data in serous ovarian cancer. Using the specific endothelial marker CD34, microvessel determinations were quantified in 132 serous ovarian tumors by manual counting at final magnifications of x 200 and x 400 in the most highly vascular areas. Computer-assisted image analysis microvessel counts, endothelial area estimates, and minimum spanning tree (MST) analysis of capillary architecture, which involves assessment of intercapillary distances, were correlated with traditional manual techniques.Manual, semiautomated, and advanced image analysis methods were found to be highly reproducible and express strong correlation with one another. Univariate cyclooxygenase analysis revealed angiogenesis parameters to be highly significant predictors for overall survival (OS) and disease-free survival. Multivariate cyclooxygenase analysis revealed maximum MST (P = 0.009), length MST (P = 0.005), 1 nearest neighbor (P