Artificial intelligence estimates the impact of human papillomavirus types in influencing the risk of cervical dysplasia recurrence: progress toward a more personalized approach

Treatment characteristics, HPV genotype distribution and risk of subsequent disease among women with high-grade cervical intraepithelial neoplasia in Europe: A systematic literature review

INTRODUCTION

High-grade cervical intraepithelial neoplasia (CIN), a premalignant lesion of the uterine cervix, is caused by persistent Human Papillomavirus (HPV) infection. CIN can be identified through screening programs and high-grade CIN is usually treated by ablation or excision. This study aimed to summarize the clinical management and outcomes among women with high-grade CIN in Europe.

METHODS

A systematic literature review was conducted to identify treatment methods and their frequency of use, report HPV genotype prevalence and distribution and summarize patterns for subsequent lesions after primary treatment, among women with high-grade CIN in Europe. Embase®, MEDLINE® and Cochrane databases were searched (1st January 2012 to 30th August 2022), along with relevant conference proceedings (2018-2022), inclusive. Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) compliant methodology was adopted. Eligibility criteria included non-immunosuppressed female patients with CIN2+ from Europe (published in English).

RESULTS

In total, n = 55 studies were included. CIN excisional therapy was the most received treatment approach (15.8-100 %, loop electrosurgical excision procedure/large loop excision of the transformation zone most common), followed by ablative therapies (1-43.3 %, cold coagulation most common). Other approaches included 'wait and watch' (4.8-52.6 %) and hysterectomy (4.8-16.2 %). HPV positivity rates ranged from 67.8-100 % pre-conization and 4.7-32.8 % post-conization. The most prevalent HPV genotypes reported (both pre- and post-treatment) were HPV16 and HPV18. In patients who received excisional or ablative procedures subsequent CIN was most frequently diagnosed ≤6 months after treatment. The overall rate of subsequent CIN reported was 0.5-20.9 %.

CONCLUSION

Conization and ablation were the most common techniques, however, these procedures were associated with sub-optimal outcomes. Close clinical follow-up is important due to the risk of subsequent CIN or invasive cancer. This review serves as a reference point for the comparison of future treatment patterns as they evolve across Europe, following improved implementation of prophylactic HPV vaccination and screening.

Using a machine learning algorithm to predict outcome of primary cytoreductive surgery in advanced ovarian cancer

OBJECTIVE

To develop a machine learning (ML) algorithm to predict outcome of primary cytoreductive surgery (PCS) in patients with advanced ovarian cancer (AOC) METHODS: This retrospective cohort study included patients with AOC undergoing PCS between January 2017 and February 2021. Using radiologic criteria, patient factors (age, CA-125, performance status, BRCA) and surgical complexity scores, we trained a random forest model to predict the dichotomous outcome of optimal cytoreduction (<1 cm) and no gross residual (RD = 0 mm) using JMP-Pro 15 (SAS). This model is available at https://ipm-ml.ccm.sickkids.ca.

RESULTS

One hundred and fifty-one patients underwent PCS and randomly assigned to train (n = 92), validate (n = 30), or test (n = 29) the model. The median age was 58 (27-83). Patients with suboptimal cytoreduction were more likely to have an Eastern Cooperative Oncology Group 3-4 (11% vs. 0.75%, p = 0.004), lower albumin (38 vs. 41, p = 0.02), and higher CA125 (1126 vs. 388, p = 0.012) than patients with optimal cytoreduction (n = 133). There were no significant differences in age, histology, stage, or BRCA status between groups. The bootstrap random forest model had AUCs of 99.8% (training), 89.6%(validation), and 89.0% (test). The top five contributors were CA125, albumin, diaphragmatic disease, age, and ascites. For RD = 0 mm, the AUCs were 94.4%, 52%, and 84%, respectively.

CONCLUSION

Our ML algorithm demonstrated high accuracy in predicting optimal cytoreduction in patients with AOC selected for PCS and may assist decision-making.

The Use of Artificial Intelligence for Complete Cytoreduction Prediction in Epithelial Ovarian Cancer: A Narrative Review

INTRODUCTION

In patients affected by epithelial ovarian cancer (EOC) complete cytoreduction (CC) has been associated with higher survival outcomes. Artificial intelligence (AI) systems have proved clinical benefice in different areas of healthcare.

OBJECTIVE

To systematically assemble and analyze the available literature on the use of AI in patients affected by EOC to evaluate its applicability to predict CC compared to traditional statistics.

MATERIAL AND METHODS

Data search was carried out through PubMed, Scopus, Ovid MEDLINE, Cochrane Library, EMBASE, international congresses and clinical trials. The main search terms were: Artificial Intelligence AND surgery/cytoreduction AND ovarian cancer. Two authors independently performed the search by October 2022 and evaluated the eligibility criteria. Studies were included when data about Artificial Intelligence and methodological data were detailed.

RESULTS

A total of 1899 cases were analyzed. Survival data were reported in 2 articles: 92% of 5-years overall survival (OS) and 73% of 2-years OS. The median area under the curve (AUC) resulted 0,62. The model accuracy for surgical resection reported in two articles reported was 77,7% and 65,8% respectively while the median AUC was 0,81. On average 8 variables were inserted in the algorithms. The most used parameters were age and Ca125.

DISCUSSION

AI revealed greater accuracy compared against the logistic regression models data. Survival predictive accuracy and AUC were lower for advanced ovarian cancers. One study analyzed the importance of factors predicting CC in recurrent epithelial ovarian cancer and disease free interval, retroperitoneal recurrence, residual disease at primary surgery and stage represented the main influencing factors. Surgical Complexity Scores resulted to be more useful in the algorithms than pre-operating imaging.

CONCLUSION

AI showed better prognostic accuracy if compared to conventional algorithms. However further studies are needed to compare the impact of different AI methods and variables and to provide survival informations.

Assessing the Long-Term Role of Vaccination against HPV after Loop Electrosurgical Excision Procedure (LEEP): A Propensity-Score Matched Comparison

Background: Primary prevention through vaccination is a prophylactic approach aiming to reduce the risk of developing human papillomavirus (HPV)-related lesions. No mature and long-term data supported the adoption of vaccination in women undergoing conization. Methods: This is a retrospective multi-institutional study. Charts of consecutive patients undergoing conization between 2010 and 2014 were collected. All patients included had at least 5 years of follow-up. We compared outcomes of patients undergoing conization plus vaccination and conization alone. A propensity-score matching algorithm was applied in order to reduce allocation biases. The risk of developing recurrence was estimated using Kaplan-Meir and Cox hazard models. Results: Overall, charts of 1914 women were analyzed. The study group included 116 (6.1%) and 1798 (93.9%) women undergoing conization plus vaccination and conization alone, respectively. Five-year recurrence rate was 1.7% (n = 2) and 5.7% (n = 102) after conization plus vaccination and conization alone, respectively (p = 0.068). After the application of a propensity-score matching, we selected 100 patients undergoing conization plus vaccination and 200 patients undergoing conization alone. The crude number of recurrences was 2 (2%) and 11 (5.5%) for patients undergoing conization plus vaccination and conization alone, respectively (p = 0.231). Vaccination had no impact on persistent lesions (no negative examination between conization and new cervical dysplasia; p = 0.603), but reduced the risk of recurrent disease (patients who had at least one negative examination between conization and the diagnosis of recurrent cervical dysplasia; p = 0.031). Conclusions: Patients having vaccination experience a slightly lower risk of recurrence than women who had not, although not statistically significantly different. Further evidence is needed to assess the cost effectiveness of adopting vaccination in this setting.

The impact of HPV-specific infection in women diagnosed with atypical glandular cells: Results from the HPV-AGC study

OBJECTIVE

To evaluate the impact of various HPV types on the risk of developing lesions of the uterus (either uterine cervix and endometrium) in women diagnosed with "atypical glandular cells" (AGC) at Pap smear.

METHODS

This is a multi-institutional retrospective study. Data of women diagnosed with AGC were retrospectively reviewed. All patients included had data about HPV DNA testing and 1-year clinical follow-up.

RESULTS

Overall, chart of 480 patients were evaluated. After the exclusion of 286 patients, data of 194 patients were available for the analysis. Mean age was 43.9 (±6.0) years. HPV infection was documented in 136 women (70.1 %). Among HPV positive patients the risk of having/developing a lesion was 33.8 % (n = 46). Lesions included low- (L-SIL) and high- (H-SIL) squamous intraepithelial lesions, in situ adenocarcinoma of the uterine cervix, invasive cancer of the uterine cervix, endometrial hyperplasia and endometrial cancer in 16 (11.7 %), 18 (13.2 %), 6 (4.4 %), 3 (2.2 %), 2 (1.5 %) and 1 (1%), respectively. Among HPV negative patients the risk of having/developing a lesion was 15.5 %. They included l-SIL, H-SIL, in situ adenocarcinoma, endometrial hyperplasia and endometrial cancer in 1 (1.7 %), 1 (1.7 %), 1 (1.7 %), 3 (5.1 %) and 3 (5.1 %), respectively. Patients diagnosed with HPV16 were at higher risk of having/developing cervical lesions in comparison to patients with other HPV infections (p < 0.01). In comparison to other HPV types, the presence of HPV 18, 31, 33, and 45 did not increase the risk of developing a lesion over the time (p > 0.2). HPV positive patients were at higher risk of being diagnosed with a cervical lesion within 6 months from detection of AGC.

CONCLUSIONS

Patients diagnosed with AGC are at risk to have / developing cervical and uterine lesions. Further prospective evidence is needed.

Age-specific predictors of cervical dysplasia recurrence after primary conization: analysis of 3,212 women

Objective

This study aimed to identify predictors of recurrence/persistence of cervical intraepithelial neoplasia grade 2+ (CIN2+) lesion (r-CIN2+) after primary conization.

Methods

Retrospective analysis involving all consecutive women having conization for CIN2+ between 1998 and 2018. The risk of r-CIN2+ was assessed using Kaplan-Meier and Cox models.

Results

Data of 3,212 women were retrospectively identified. After a mean follow-up of 47 (±22.2) months, 112 (3.5%) patients developed r-CIN2+. Mean time interval between prior conization and diagnosis of r-CIN2+ was 26.2 (±13.2) months. Via multivariate analysis, presence of high-risk human papillomavirus (HPV) types at the time of CIN2+ diagnosis, hazard ratio (HR)=3.40 (95% confidence interval [CI]=1.66–6.95) for HPV16/18 and HR=2.59 (95% CI=1.21–5.55) for HPV types other than 16/18, positive margins at primary conization, HR=4.11 (95% CI=2.04–8.26) and HPV persistence after conization, HR=16.69 (95% CI=8.20–33.9), correlated with r-CIN2+, independently. Considering age-specific HPV types distribution, we observed that HPV16/18 infection correlated to an increased risk of r-CIN2+ only in young women (aged ≤25 years; p=0.031, log-rank test); while in the older population (>25 years) HPV type(s) involved had not impact on r-CIN2+ risk (p>0.200, log-rank test).

Conclusion

HPV persistence is the main factor predicting r-CIN2+. Infection from HPV16/18 has a detrimental effect in young women, thus highlighting the need of implementing vaccination against HPV in this population. Further prospective studies are warranted for tailoring clinical decision-making for post-conization follow-up on the basis of risk factors.

Population-based outcomes by immunosuppressed status in patients undergoing radiotherapy for oropharyngeal cancer

INTRODUCTION

The incidence of immunosuppression in patients with oropharynx head & neck squamous cell carcinoma (SCC) is not well studied. This study evaluates disease characteristics and treatment outcomes in oropharynx SCC in patients with and without immunosuppression.

METHODS

A retrospective review of all patients treated with radiotherapy for oropharynx SCC at BC Cancer from 2011 to 2016 was performed. Survival outcomes were assessed using Kaplan-Meier methods and competing risk analysis. Multivariate analysis and propensity score matching were performed.

RESULTS

There were 1077 patients, of which 5.8% (n = 62) had an immunosuppressive medical condition or were taking long-term immunosuppressive medication at diagnosis. Median follow-up was 3.3 years. Three year OS for patients without immunosuppression was 79.5% (95% Confidence Interval [CI] 76.8-82.0%) and for those with immunosuppression was 64.6% (95% CI 50.9-75.3%) (hazard ratio [HR] 1.78, 95% CI 1.18-2.68, p = 0.0062). The three year disease recurrence for patients without immunosuppression was 24.9% (95% CI 22.2-27.7%) and 44.4% (95% CI 31.5-56.6%) for those with immunosuppression (HR 2.12, 95% CI 1.45-3.11, p = 0.0001). Multivariate analysis of disease free survival (DFS) found that active smoking, advanced TNM stage, base of tongue subsite, p16 negative and unknown, no concurrent chemotherapy, higher Charlson Comorbidity Index, and lower radiation dose were also associated with worse DFS (all p < 0.05). Immunosuppressed patients had worse DFS relative to patients without immunosuppression, p < 0.001, HR 1.97 (95% CI 1.33-2.91).

CONCLUSION

Immunosuppression was an independent predictor of worse DFS in this large cohort of patients with oropharynx SCC.

Artificial Intelligence Estimates the Importance of Baseline Factors in Predicting Response to Anti-PD1 in Metastatic Melanoma

OBJECTIVE

Prognosis of patients with metastatic melanoma has dramatically improved over recent years because of the advent of antibodies targeting programmed cell death protein-1 (PD1). However, the response rate is ~40% and baseline biomarkers for the outcome are yet to be identified. Here, we aimed to determine whether artificial intelligence might be useful in weighting the importance of baseline variables in predicting response to anti-PD1.

METHODS

This is a retrospective study evaluating 173 patients receiving anti-PD1 for melanoma. Using an artificial neuronal network analysis, the importance of different variables was estimated and used in predicting response rate and overall survival.

RESULTS

After a mean follow-up of 12.8 (±11.9) months, disease control rate was 51%. Using artificial neuronal network, we observed that 3 factors predicted response to anti-PD1: neutrophil-to-lymphocyte ratio (NLR) (importance: 0.195), presence of ≥3 metastatic sites (importance: 0.156), and baseline lactate dehydrogenase (LDH) > upper limit of normal (importance: 0.154). Looking at connections between different covariates and overall survival, the most important variables influencing survival were: presence of ≥3 metastatic sites (importance: 0.202), age (importance: 0.189), NLR (importance: 0.164), site of primary melanoma (cutaneous vs. noncutaneous) (importance: 0.112), and LDH > upper limit of normal (importance: 0.108).

CONCLUSIONS

NLR, presence of ≥3 metastatic sites, LDH levels, age, and site of primary melanoma are important baseline factors influencing response and survival. Further studies are warranted to estimate a model to drive the choice to administered anti-PD1 treatments in patients with melanoma.

Cervical intraepithelial neoplasia in women who had vaccination against HPV

OBJECTIVE

To investigate the characteristics of women developing cervical intraepithelial neoplasia who had had a vaccination against human papillomavirus (HPV).

METHODS

A retrospective cohort study was carried out of women diagnosed with moderate or severe cervical dysplasia (CIN2+) in four Italian centers between 2015 and 2017. All women included had had previous bivalent or quadrivalent vaccination against HPV.

RESULTS

The present study included 43 patients affected by CIN2+. The median age was 28 (range, 21-41) years. Ten (23.3%) patients did not have a diagnosis of specific HPV type(s) involved: high-risk HPV was detected in 7 (16.3%) women while HPV testing was negative in 3 (6.9%) women. Lesions related to HPV16 were found in two patients. HPV types covered by nonavalent vaccination were diagnosed in 27/33 (81.8%) women. HPV types not covered by nonavalent vaccination were diagnosed in 6 (18.2%) women. Co-infections are most commonly detected in women with HPVs other than those included in the nonavalent vaccination (P=0.024).

CONCLUSION

Cervical dysplasia occurring after HPV vaccination is a rare condition. Theoretically, nonavalent vaccination should improve protection against more than 80% of HPV-related lesions compared to other vaccines.

Screening of Cervical Cancer with Self-Collected Cervical Samples and Next-Generation Sequencing

Cervical cancer is the second leading cause of death in female genital malignancies. Persistent infection with high-risk HPV is closely related to cervical intraepithelial neoplasia (CIN). Wide-scale HPV screening has already been implemented in developed countries. However, with advances in HPV testing methods, there are presently no better methods for the management of the increasing number of high-risk HPV-positive women except for periodic review. In order to improve screening coverage and achieve better triage of those women, we present current HPV testing methods with self-collected cervical samples, focusing on recent advances in next-generation sequencing (NGS) technologies as a promising screening technology for cervical cancer precursors.

Treatment modalities for recurrent high-grade vaginal intraepithelial neoplasia

Objective

We have investigated outcomes of women presenting with recurrent high-grade vaginal intra-epithelial neoplasia.

Methods

Data of consecutive women diagnosed with recurrent high-grade vaginal intra-epithelial neoplasia after primary treatment(s) were retrieved. Risk of developing new recurrence over the time was assessed using Kaplan-Meier and Cox models.

Results

Data of 117 women were available for the analysis. At primary diagnosis, 41 (35%), 4 (3.4%) and 72 (61.6%) patients had had laser, pure surgical and medical treatments, respectively. Secondary treatments included: laser ablation and medical treatment in 95 (81.2%) and 22 (18.8%) cases, respectively. After a mean (standard deviation) follow-up of 72.3 (±39.5) months, 37 (31.6%) out of the entire cohort of 117 patients developed a second recurrence. Median time to recurrence was 20 (range,5-42) months. Patients with recurrent high-grade vaginal intra-epithelial neoplasia undergoing medical treatments were at higher risk of developing a second recurrence in comparison to women having laser treatment (p=0.013, log-rank test). After we corrected our results for type of treatment used for recurrent disease, we observed that the execution of primary laser treatment was independently associated with a lower risk of developing new recurrences (hazard ratio [HR]=0.46; 95% confidence interval [CI]=0.21–0.99; p=0.050). The other variable that is independently associated with a new recurrence is the persistent infection from HPV16 or 18 (HR=3.87; 95% CI=1.15–13.0; p=0.028).

Conclusion

Patients with recurrent high-grade vaginal intra-epithelial neoplasia are at high risk of developing new recurrences. Our data underline that the choice of primary treatment might have an impact of further outcomes.

Artificial intelligence weights the importance of factors predicting complete cytoreduction at secondary cytoreductive surgery for recurrent ovarian cancer

Objective

Accumulating evidence support that complete cytoreduction (CC) at the time of secondary cytoreductive surgery (SCS) improves survival in patients affected by recurrent ovarian cancer (ROC). Here, we aimed to determine whether artificial intelligence (AI) might be useful in weighting the importance of clinical variables predicting CC and survival.

Methods

This is a retrospective study evaluating 194 patients having SCS for ROC. Using artificial neuronal network (ANN) analysis was estimated the importance of different variables, used in predicting CC and survival. ANN simulates a biological neuronal system. Like neurons, ANN acquires knowledge through a learning-phase process and allows weighting the importance of covariates, thus establishing how much a variable influences a multifactor phenomenon.

Results

Overall, 82.9% of patients had CC at the time of SCS. Using ANN, we observed that the 3 main factors driving the ability of achieve CC included: disease-free interval (DFI) (importance: 0.231), retroperitoneal recurrence (importance: 0.178), residual disease at primary surgical treatment (importance: 0.138), and International Federation of Gynecology and Obstetrics (FIGO) stage at presentation (importance: 0.088). Looking at connections between different covariates and overall survival (OS), we observed that DFI is the most important variable influencing OS (importance: 0.306). Other important variables included: CC (importance: 0.217), and FIGO stage at presentation (importance: 0.100).

Conclusion

According to our results, DFI should be considered as the most important factor predicting both CC and OS. Further studies are needed to estimate the clinical utility of AI in providing help in decision making process.