In this follow-up study after cervical conization, 90% of the women had a follow-up cervical sample taken and 76% of these had a normal test of cure (cervical negative HPV and normal cytology) within 12 months. A negative cervical HPV sample as a marker for healed HPV infection was found in 80% within eight months. The risk of developing a new HSIL/AIS after a negative test of cure after conization was 79/100 000 people-years, which was lower than the incidence of HSIL/AIS in the total population in Sweden where the risk was more than 270/100 000 women above 22 years of age per year in 2020. No cases of cervical cancer were found later in the follow-up time in treated women with a negative test of cure within 12 months post-conization.

This study found that 84.4% of women undergoing conization completed a test of cure within 12 months following conization. It has not previously been known at what rate women in Sweden or the region of Skåne are lost to follow-up. The percentage of women submitting a cervical sample reached close to 90%, indicating that the follow-up rate was higher for the cytological test than the test of cure. This discrepancy may be due to changed recommendations during the first years of the study. Even though the follow-up recommendations after conization have been to have a double test with HPV and cytology since 2010 in the region of Skåne, the adherence to the guidelines was inconsistent prior to the establishment of new national guidelines in 2015. It is possible that during the transition time into the new guidelines, some clinics only performed the cytological test as a follow-up. Overall, a compliance to the follow-up program of 90% is considered good and is in concordance with the results of a Danish study by Bruhn et al. [7].

The results of clearance of cervical dysplasia and HPV reveal a steady increase in the percentage of negative tests of cure between the timepoints measured, 69.7%, 76.3% and 84.4% at eight, 12 and 24 months respectively. The clearance of HPV follows the same pattern with a clearance rate of 79.6 at eight months post-conization followed by 83.2% and 87.8% at 12 and 24 months post-conization respectively. In a systematic review, the clearance of HPV after LEEP ranged from 76 to 100% at two to 35 months post-conization [13]. In another smaller study on surveillance post-conization the clearance rate of HPV was 55% at six months post-conization [14]. A Korean study investigated the clearance rate of HPV at three, six, nine, 12, 18 and 24 months post-conization and found a clearance rate of 54.4%, 85.7%, 93.7%, 97.8%, 98.5%, and 98.9% respectively. The clearance rates did not significant differ by age, parity, or severity of the cervical lesion. But HPV DNA load was found to have a significant impact on the rate of HPV clearance, with slower clearance with HPV DNA loads > 500 RLU/PC [15].

The recurrence rate of HSIL or AIS in women with a negative test of cure 12 months post-conization was 0.2% in this study, which is low and similar to other studies. In an Italian study of 310 women treated with LEEP for CIN2 + , none were diagnosed with residual/recurrent CIN2 + among women with negative HPV six months post-conization during a follow-up time of five years [14]. In a systematic review of post-treatment surveillance, the risk of CIN2 + was 0.69% after a negative HPV test and 1.7% after a negative cytological sample, with a follow-up time ranging from 24 to 36 months [16]. The results of a follow-up study in Sweden with co-testing six to 12 months after conization showed that, if the co-test was normal, none of the women had recurring CIN2 lesions within three years [11]. There has been an interest in whether an HPV test can be used on a standalone basis at the post-treatment follow-up. In a recent Danish study, it was suggested that HPV testing can be used alone as test of cure if stratifying for resection margin status [7]. In the systematic review mentioned above, a meta-analysis was made pooling studies evaluating co-testing with HPV and cytology versus HPV or cytology testing alone post-conization. The risk of CIN2 + was 0.68% among individuals with negative co-testing and 1.4% and 2.5% respectively among individuals with negative HPV and cytology results [16]. In Sweden, the consensus is to use co-testing of HPV and cytology at post-conization follow-up [2]. No cases of cervical cancer could be identified after a negative test of cure at 12 months in our study with a mean follow-up time of 32.1 months, indicating that conization combined with a follow-up test of cure is a good method for preventing and finding cervical cancer. The mean time from conization to a new diagnosis of HSIL/AIS was 1267 days, which approximates to about 3.5 years before a new cervical dysplasia evolved. It is fair to assume that if the follow-up time in this study was longer, more cases of cervical dysplasia, and eventually cervical cancer, would have occurred. A previously published study before the use of HPV analyses assessing time to new high-grade dysplasia and cancer after conization suggested that the mean time between treatment and a new diagnosis of dysplasia was five years and eight months [17].

Currently, the standard time for post-treatment follow-up in Sweden is six months post-conization [2]. In this study, the highest clearance rate for HPV infection was 24 months post-conization, but the clearance rate was high already at eight months post-conization. Comparing the HPV clearance at eight and 12 months post-conization reveals no statistically significant difference (p = 0.09) between the two timepoints. When considering the HPV results in combination with cytology results, a significant decrease in positive co-tests was seen between 12 and 24 months post-conization (p < 0.001). It is important to determine a timepoint where most women have had time to heal their HPV infection, but also and most importantly, that not enough time has passed for women with persistent infection to have developed a new HSIL or even progressed to cancer. Among women with abnormal tests of cures in this study, two women were diagnosed with cervical cancer eight months post-conization. This, in combination with a high HPV clearance rate after eight months, makes it reasonable to suggest that the current recommendation of post-treatment follow-up after six months should remain.

To our knowledge, this study is the largest post-treatment study published to date, including 8721 women with a total of more than 23 000 years of follow-up [16]. The mean time of follow-up in this study was 32.1 months, which is in accordance with other similar studies where the time of follow-up mainly ranged between 24 and 36 months [16], suggesting that the results in this study are comparable to other studies evaluating the rate at which new cases of HSIL/AIS or cancer arise after a negative test of cure. A limitation of the study is that the results are solely based on data from the LIMS database and the histopathological or cytological diagnoses have not been re-reviewed for this study. A quality re-review made in Sweden, published in 2021, revealed that among women who developed HSIL/AIS or cancer after a normal cytological test, 10–45% had falsely negative cytological tests when re-reviewed [18]. Although it is possible false negative cytological tests appeared in this study as well, all new cancer cases would have been found during the follow-up period in the database. Another limitation of the study concerns the patients lost to follow-up. In the data analyzed, it is not known whether patients are lost to follow-up due to lack of compliance or because they have migrated out of the region of Skåne and are followed up elsewhere. We do not have the data necessary to conclude if some patients attended a follow-up elsewhere. The resection margins status was not included in this study, which could be a limitation. A meta-analysis by Zhang et al. found surgical margin to be positively correlated with persistent HPV infection post-conization [20]. In another meta-analysis by Arbyn et al., analyzing incomplete excision of cervical precancer, it was found that the risk of residual or recurrent CIN2 + was greater among patients with positive resection margins, but testing for hr-HPV at the post-treatment follow-up was a more accurate predictor of treatment failure than margin status [19].