Introduction

Endometriosis is a complex, estrogen-dependent inflammatory disorder that affects approximately 10% of women of reproductive age1. Characterized by pelvic pain, infertility, and fatigue, it has a profound impact on patients’ physical, reproductive, and emotional health2. Despite its high prevalence, there is no definitive or causal treatment for endometriosis, as the pathophysiology of the disease remains poorly understood. As a result, endometriosis is considered a chronic condition that requires long-term, multidisciplinary management.

Current therapeutic approaches are symptomatic and include medical treatments (mainly nonsteroidal anti-inflammatory drugs and hormonal suppression) and/or surgical removal of lesions. Since both hormonal and surgical therapies seem to achieve comparable pain improvement overall3, there has been a growing tendency towards a less invasive management, reserving laparoscopic resection for patients who decline or fail to respond to medical treatments, in cases involving large endometriomas (>5 cm), or when ultrasound findings suggest malignancy or risk of obstruction in the urinary or intestinal tract, among others4,5,6.

While surgical resection of endometriosis lesions can provide symptom relief for many patients, postoperative recurrence is common, with pain relapse generally reported in 20–26% of patients within the first year after surgery7,8,9, and lesion reappearance rates ranging from 12 to 29%, 1 and 2 years postoperatively, respectively7,9,10. The term postoperative recurrence is generally used across the literature to refer to both events indistinctively, which adds to the variability of reported outcomes. When integrating these heterogeneous definitions, the estimated mean 2-year recurrence of 21.5%11.

The high rates of postoperative recurrence of both pain symptoms and endometriosis lesions are important contributors to the complexity of the long-term management of endometriosis11,12. In fact, recurrence has been identified as a top research priority by both patients and healthcare professionals, who call for more effective strategies to prevent disease progression following treatment13,14.

Although this issue is widely acknowledged in clinical practice, it has received little attention and is frequently framed as surgical failure, rather than as a biological phenomenon worth investigating. As a result, the mechanisms that drive disease recurrence remain poorly understood. To improve long-term outcomes, it is essential to understand why endometriosis lesions and symptoms reappear in some individuals despite seemingly adequate treatment, and what biological factors underlie such predisposition.

This review aims to (1) provide an updated and integrative synthesis of the literature on endometriosis recurrence definitions, rates, and risk factors; (2) explore the biological processes that may contribute to lesion recurrence, in light of current knowledge on the pathogenesis of endometriosis; and (3) propose recurrent endometriosis as a conceptual model to gain insights into disease mechanisms, while outlining future directions for translational research.

Conceptual and clinical heterogeneity in endometriosis recurrence definitions

Despite being an important concern in endometriosis care, there is currently no standardized definition of postoperative recurrence. In a conceptual sense, recurrence refers to the return of disease after a period of remission. In endometriosis, recurrence can describe two distinct phenomena: the detection of endometriotic lesions after complete surgical resection, which represents the reappearance of the disease itself and is considered the most objective and biologically informative endpoint, or the return of endometriosis-associated pain symptoms, which is more relevant to patient management.

Although they often overlap, lesion regrowth and symptom relapse are not always equivalent15,16,17. This lack of correlation is generally reflected in the recurrent rates reported for both outcomes, where pain recurrence is typically higher than lesion recurrence. For example, in a retrospective study of 115 patients with deep infiltrating endometriosis (DIE), relapse of pain 3 and 5 years after surgery was 20.5 and 43.5%, and the lesion recurrence rates were 9 and 28%, respectively18. Similarly, another retrospective study of 401 patients reported pain recurrence at 34% and lesion recurrence detected by transvaginal ultrasonography (TVUS) at 16%, both within a median follow-up of 24 months19.

This highlights that endometriosis-associated symptoms can be driven by pathways unrelated to anatomical lesion regrowth, such as central sensitization, pain comorbidities, or factors secondary to the surgical procedure itself, including fibrosis or the formation of new adhesions1,20,21. Central sensitization refers to an increased excitability of central pain pathways, leading to pain amplification and persistence even in the absence of peripheral noxious stimuli. This mechanism has been associated with poor response to surgery and hormonal therapies, and may contribute to postoperative pain recurrence in the absence of visible disease20. Additionally, persistent symptoms can be related to undetected peritoneal implants not removed during surgery22. It is also worth noting that studies often fail to differentiate between symptom persistence (indicating initial treatment failure) and relapse (the return of pain after a period of relief, potentially linked), further complicating the understanding of these outcomes from a pathophysiological standpoint.

Conversely, lesion reappearance does not always entail symptom relapse: retrospective studies on recurrent endometriomas have reported that 24-30% of patients with imaging-confirmed recurrence remained asymptomatic19,23. Therefore, recurrence should be considered a multidimensional phenomenon involving both lesion regrowth and symptom reappearance, which likely involve distinct underlying mechanisms24.

Measuring recurrence in endometriosis: proxies and methodological variability

The lack of consensus in defining, assessing, and consequently reporting recurrence is a persistent methodological challenge frequently highlighted by systematic reviews and meta-analyses25,26,27. Notably, Ceccaroni et al.28 systematically compiled the definitions used across observational studies and clinical trials, illustrating this heterogeneity and the limited comparability across studies. There is currently no commonly agreed, well-defined, and consistently applied set of outcomes to evaluate postoperative recurrence in endometriosis. Instead, a diverse range of clinical and imaging-based assessment criteria are used, each with its own limitations.

Imaging modalities such as TVUS and magnetic resonance imaging (MRI) are the most widely used tools to assess lesion reappearance. As in the diagnostic process, the accuracy of these tools depends on subtype, lesion localization and size, and operator expertise29. Since both TVUS and MRI are highly reliable in detecting ovarian endometriosis, it is the most studied subtype in terms of postoperative recurrence, although specific cyst size cut-offs and sonographic features remain variable between studies27,30,31. A recent meta-analysis by Veth et al.31 has demonstrated a progressive increase in endometrioma recurrence risk over time, with rates of 4% at 3 months post-surgery, 14% at 6 months, 17% at 17 months, and 27% at 24 months when not using postoperative hormonal treatment.

However, as in diagnosis, a negative imaging result does not necessarily rule out the presence of disease, so histological confirmation of endometriotic lesions is the most rigorous approach to confirm lesion recurrence, although highly invasive and unsuitable for longitudinal assessment. Additionally, repeated surgeries are generally discouraged due to the associated loss of ovarian reserve, potential surgical complications, and the formation of new adhesions4,32,33,34. Consequently, not all recurrence cases are managed surgically, and recurrence rates might be underestimated.

Beyond imaging, the longitudinal assessment of endometriosis recurrence is further limited by the absence of a reliable, non-invasive biomarker that could serve as a dynamic indicator of disease activity over time. In a chronic condition like endometriosis, where repeated surgical verification is neither feasible nor ethical, such a biomarker would be key to monitoring recurrence and disease progression.

Symptom recurrence is usually assessed through patient-reported outcomes, primarily pain scores. While these are clinically meaningful, they are inherently subjective and prone to reporting and attrition bias, especially in long-term follow-up studies7. Definitions vary widely: some studies dichotomize pain as “present” or “absent,” while others use changes in visual analog scale scores or relative improvement thresholds35.

Both lesion and symptom recurrence can prompt subsequent changes in clinical management, and many retrospective studies rely on reoperation or starting a second-line hormonal therapy as indirect indicators of recurrence25,36,37. Although practical, these proxies cannot distinguish between treatment failure and true recurrence after remission and usually do not confirm the presence of lesions histologically, so they should be interpreted carefully.

The inconsistent use of diverse outcomes to assess recurrence across studies severely limits comparability and complicates the interpretation of recurrence rates, risk factors, and therapeutic outcomes. Moreover, recurrence estimates are influenced by cohort characteristics, disease subtype and severity, surgical technique, follow-up length, and the use (or absence) of postoperative hormonal therapy. Collectively, this variability not only explains the wide range of recurrence rates reported in the literature11 but also limits our understanding of the biological and clinical dynamics of recurrence.

Risk factors for lesion and symptom recurrence

Understanding which factors predispose patients to postoperative recurrence is essential to identify patients at higher risk of recurrence, to ultimately guide individualized management. However, as emphasized above, the substantial heterogeneity in how recurrence is defined across the literature complicates the interpretation and comparison of findings. Table 1 summarizes the main clinical, surgical, and patient-related determinants influencing the likelihood of recurrence reported in the literature, distinguishing between risk factors for lesion and symptom recurrence, and including studies using broader definitions as well. Because imaging reliably detects ovarian endometriomas, it has naturally become the most extensively studied subtype in terms of recurrence, and this predominance should be considered when interpreting risk factors.

Table 1 Risk factors for endometriosis postoperative recurrence
Full size table

Nevertheless, several consistent patterns can be identified across studies. Younger age at the time of surgery is associated with a higher risk of recurrence38. Although the absolute differences in mean age between recurrent and non-recurrent groups are small, this consistent trend9,18,19,39,40,41,42,43,44,45,46,47,48,49,50,51 suggests a biologically meaningful effect likely related to longer lifetime estrogen exposure. Preoperative pain symptoms, including dysmenorrhea and chronic pelvic pain, have been linked to both lesion reappearance19,43,52,53,54,55,56 and pain persistence and recurrence19,47,57, possibly reflecting pre-existing inflammatory and nociplastic factors not resolved by surgery.

Disease-related factors such as extensive adhesions49,54,58,59, larger cysts42,54,55,60,61,62,63,64,65, coexistent deep infiltrating lesions61,66 and adenomyosis47,66 or higher preoperative revised American Society for Reproductive Medicine (rASRM) scores18,19,24,39,41,42,48,50,58,59,67,68,69,70 are associated with increased lesion recurrence risk mainly, likely due to incomplete lesion resection.

Surgical approach and postoperative care also impact outcomes. Conservative surgery is widely associated with earlier recurrence25,41,48,52,58,71,72,73,74. Conversely, the use of postoperative hormonal therapy has been consistently found as a protective factor against recurrence9,26,50,69,70,75,76,77. However, an insufficient duration or poor adherence to treatment over the long term can diminish the protective effects of these therapies10,44,61,74,78,79,80,81,82. Similarly, achieving pregnancy after surgery appears to lower recurrence risk, likely due to prolonged ovarian suppression19,40,41,44,48,51,56,59,63,67,70,81,83.

The role of postoperative hormonal treatment in recurrence prevention and management

Among the risk factors for endometriosis recurrence mentioned above, periods of ovarian suppression or amenorrhea, whether pharmacologically induced or occurring during pregnancy and lactation, are associated with a reduced risk of both lesion and symptom recurrence. In this hypo-estrogenic environment, the growth of residual lesions is suppressed, new lesion development is prevented, and associated pain symptoms persisting or reappearing after surgery are controlled.

Postoperative medical therapy represents a cornerstone in endometriosis recurrence prevention. Systematic reviews and meta-analyses have reported that the risk of lesion reappearance is reduced by 59–70%, and symptom recurrence by 30% after 12 months, if using hormonal treatment postoperatively, compared to surgery alone7,26. Comparative studies assessing different hormonal agents and regimens for recurrence prevention have reported heterogeneous results26,60,79,82,84,85,86,87. Variability in study design, inclusion criteria, treatment duration, adherence, and follow-up complicates comparisons and limits definitive conclusions. Overall, no hormonal agent has been consistently demonstrated to be better than others, suggesting that the protective effect primarily derives from sustained ovarian suppression rather than from a specific drug class. In line with this, a randomized controlled trial by Seracchioli et al.10 demonstrated that continuous oral contraceptive use was associated with lower rates of dysmenorrhea relapse and severity compared to cyclic use. Accordingly, the European Society of Human Reproduction and Embryology guideline on endometriosis recommends the long-term use of continuous combined oral contraceptives or levonorgestrel-releasing intrauterine systems following laparoscopy for the prevention of dysmenorrhea relapse6.

Despite its benefits, hormonal therapy is not without limitations. Symptom relapse occurs shortly after discontinuation60, and recurrence rates among patients treated for less than 12 months are similar to those of untreated individuals within 1–2 years88. This suggests a predominantly suppressive effect, requiring sustained use to maintain benefit. However, long-term use is associated with relevant side effects and is not appropriate for patients with immediate pregnancy desire, so discontinuation rates are high89,90,91. Additionally, treatment failure in this context may be amplified by the fact that patients referred to surgery may be less responsive to medical therapy as a first-line approach4,92.

Beyond adherence and tolerance, growing evidence suggests that some patients may develop resistance to postoperative hormonal therapy, particularly progestins93,94. This may reflect lesion-level progesterone resistance or, in the case of persistent pain, the predominance of centrally mediated mechanisms that are less responsive to hormonal suppression. Because hormonal therapy primarily acts on lesion-related, peripheral pain pathways, its benefit may be limited in such patients.

Therefore, long-term management of endometriosis recurrence remains an important challenge. As mentioned above, reoperation is increasingly discouraged due to the associated risks and the fact that a history of previous endometriosis-related surgeries is itself a risk factor for future recurrence51,59,65,95,96, creating a vicious cycle that further complicates long-term management97.

These challenges highlight the need to identify new preventive and therapeutic strategies to manage endometriosis recurrence, ideally beyond hormonal suppression. To develop such strategies and better tailor the long-term management of endometriosis, it is essential to better understand the biological basis of recurrence and to develop tools to identify patients at higher risk.

Predicting recurrence: clinical models and emerging biomarkers

Accurately predicting postoperative recurrence in endometriosis is key to advancing personalized care and guiding long-term treatment strategies. A limited number of predictive models have been developed based on clinical, surgical, and molecular variables, but their performance and applicability remain modest, partly due to the variable criteria used to define recurrent patients.

Most clinically based models rely on established risk factors, such as adhesions or high rASRM scores, and their use primarily relates to the risk of lesion recurrence, since they reflect the anatomical severity of endometriosis. For example, Huang et al.67 proposed a logistic regression model using rASRM >70 as a single predictor (AUC = 0.79), but its applicability is constrained by a short follow-up and a strict recurrence definition, requiring lesion recurrence (detected by TVUS and pelvic exam), high CA125 levels, and symptom return. Broader criteria are used by Su and Xie50, who combined clinical and inflammatory parameters and achieved higher accuracy (AUC = 0.895) in detecting symptomatic recurrence, supported by either high CA125 levels, pelvic exam, or TVUS findings. However, the inclusion of CA125 as a predictor in these models limits their reliability, given its low specificity and poor sensitivity in early-stage disease28,98, and considering that recent clinical guidelines discourage its use for detecting endometriosis in routine practice99. Lastly, a model by Li et al.68 showed good performance (AUC = 0.80), which still requires validation in larger cohorts and with longer follow-up durations.

Looking ahead, the growing availability of large-scale electronic health record data provides an opportunity to develop machine learning-based predictive scores for endometriosis outcomes with excellent statistical power. For example, Mustard et al.57 used national and research databases to develop predictive models of pain reduction following endometriosis surgery, identifying the removal of ovarian and uterosacral ligament endometriosis as important predictors of pain relief, although pending on external validation. Although still in early stages and focused on pain outcomes, integrating such datasets with molecular and imaging data could open promising avenues for individualized recurrence risk assessment.

In parallel, exploratory studies have identified candidate RNA and protein biomarkers differing between recurrent and non-recurrent patients in serum or endometrioma tissue (Table 2). Although promising, none of these biomarkers have yet been validated in independent cohorts, and their utility in predicting recurrence remains to be established.

Table 2 Potential biomarkers for endometriosis postoperative recurrence published in the literature
Full size table

At the RNA level, decreased circulating levels of the long non-coding RNAs UCA1100 and LINC01465101 have been linked to a higher risk of recurrence at 2–3 years post-surgery, whereas elevated serum LGMNP1 carried by extracellular vesicles in recurrent patients achieved an AUC of 0.869, and was shown to promote M2 macrophage polarization102. In lesion samples, overexpression of miR-20a, the stemness-associated gene NANOG103, and lncRNA H19 have been associated with recurrence, with H19 achieving an AUC of 0.728104. However, many of these studies did not control for important confounders like the use of postoperative therapy, which limits their interpretation.

Tissue-based immunohistochemical (IHC) biomarkers have also shown potential. Studies from Prof. Guo’s group identified differential expression of COX2, NFkB-p65, PRB, SLIT, and ROBO1 in recurrent endometriomas, which served to develop predictive models with high reported sensitivity and specificity (up to 86% and 87%, respectively) (Table 3)105,106,107. Building on these results, perioperative treatment with non-specific β-blockers or NFκB inhibitors reduced recurrence rates in mouse models, suggesting therapeutic relevance108,109. Additionally, Slit2 overexpression increased lesion size and vascularization in mice, supporting its role in endometriosis angiogenesis110. However, independent studies have reported opposite expression patterns of NFκB-p65 and PRB in recurrent lesions, likely reflecting differences in study design, marker localization (epithelial vs stromal), and patient characteristics111,112. This highlights the urgent need for harmonized protocols and replication in larger, well-characterized cohorts, ideally incorporating single-cell and spatial omics technologies to capture the cellular heterogeneity of endometriotic tissue.

Table 3 Predictive models for endometriosis postoperative recurrence published in the literature
Full size table

Additional candidates include markers of proliferation (Ki-67)113, and adhesion and extracellular matrix remodeling (E-cadherin, uPA, MMP9, and EMMPRIN)114, as well as inflammatory mediators (S100A8/A9)115, assessed in endometriomas by IHC. Although biologically plausible, these findings are preliminary, as they have been performed in small cohorts and their predictive performance remains untested.

Developing robust prognostic tools for recurrence risk prediction will be essential to stratify patients, optimize postoperative care, and improve long-term outcomes. Beyond risk prediction, validated biomarkers may also reveal biological mechanisms driving recurrence and serve as novel therapeutic targets.

The role of incomplete surgery and minimal residual disease in endometriosis lesion recurrence

The mechanisms driving the reappearance of endometriotic lesions after surgery are not fully understood and are likely multifactorial. While surgical factors are well-established contributors to recurrence, it may also result from biological processes intrinsic to the disease and its interaction with the host environment. Figure 1 presents a visual summary of the main mechanisms contributing to endometriosis lesion recurrence described in the following sections.

Fig. 1: Potential biological mechanisms of endometriosis recurrence.
Fig. 1: Potential biological mechanisms of endometriosis recurrence.
Full size image

The left panel illustrates surgery-related mechanisms contributing to lesion recurrence, including minimal residual disease (resulting from incomplete excision, invisible peritoneal lesions, or impaired immune clearance of remaining endometriotic cells as a result of surgical transient immunosuppression) as well as the unintentional dissemination of viable endometrial material with the capacity to implant in the peritoneal cavity and give rise to recurrent lesions. The right panel depicts potential mechanisms of de novo lesion formation, integrating several hypotheses described in the literature. Surgical scars are also represented as favorable implantation sites for disseminated endometriotic cells or menstrual debris, potentially facilitating lesion establishment. Created in BioRender. Delgado, R. (2026) https://BioRender.com/1s91uhu.

Incomplete surgical removal is one of the most established contributors to recurrence18,116. Minimal residual disease, whether due to conservative approaches or incomplete surgery, significantly increases the risk of recurrence across all endometriosis subtypes (reviewed in refs. 11,109,117). Factors influencing the completeness of excision, such as lesion location, disease severity, the presence of adhesions, and surgical expertise118,119 have been discussed above. And even when surgery seems macroscopically complete, residual microscopic implants may go undetected, which may partly explain why recurrent lesions often appear in the same anatomical location23,120.

This is particularly relevant in the case of DIE, where the complexity of surgery can make complete excision challenging. Some studies have identified positive resection margins in DIE as a major risk factor for early recurrence, especially when lesions are located in anatomically complex areas such as the bowel25,45,58,71,72. However, findings from Roman et al. point out that small palpable satellite nodules located near the resection site, rather than microscopic occult endometriosis at bowel resection margins, carry greater clinical relevance121,122. Surgical complications may further increase this risk72. Similarly, conservative procedures to treat ovarian endometriosis are associated with higher recurrence rates. A recently updated Cochrane review123 reported that cystectomy of endometriomas significantly reduced lesion recurrence (from 37 to 5–17%) and pain recurrence (from 49 to 10–34%) rates compared to more conservative approaches. Nevertheless, ultrasound-guided ethanol sclerotherapy has gained interest as a less invasive alternative, with recent series using standardized protocols reporting recurrence rates comparable to cystectomy124,125.

Altogether, these findings support the widely accepted interpretation that residual disease is responsible for early lesion recurrence and lesion-related pain relapse observed within months after surgery126. These minimal or invisible lesions can remain dormant for some time, especially if postoperative hormonal suppression is used briefly. Once hormonal treatment is discontinued, dormant cells may re-grow, leading to recurrence in the medium term. Indeed, while many experts support continuous hormonal suppression, it is often acknowledged that such treatments may only delay, rather than prevent, recurrence34,60,117,127.

The temporal dynamics of recurrence reinforce this view11. Recurrence commonly shows an early peak, and then enters a plateau of apparent stability, probably representing successful interventions, and rises again in the long term. However, this pattern is difficult to assess consistently in the literature due to the heterogeneity in follow-up durations and reporting, with most studies providing only short- to medium-term follow-up, typically up to 2 years, although median time to repeat surgery has been reported between 30 and 36 months39,120.

These observations suggest that recurrence due to incomplete or suboptimal surgery tends to manifest earlier, while later recurrences may reflect the slow progression of residual disease or the formation of new lesions.

Other recurrence-promoting effects of surgery

Beyond residual disease, the surgical procedure itself may contribute to lesion reappearance. Some authors have suggested that the dissemination of viable endometrial cells during surgery may facilitate the development of new lesions. For example, spillage of endometrioma cyst fluid during excision has been shown to release endometrial cells capable of implanting on surrounding tissues, potentially forming new lesions128.

Similarly, uterine trauma, such as that caused by cesarean section, has been associated with increased risk of endometriosis development129,130, and in patients with a prior diagnosis of endometriosis, with a two-fold higher risk of recurrence compared to vaginal births131. This may be a consequence of the intra-abdominal dissemination of endometrial cells131, together with the protective effect of cervical dilation and enhanced uterine drainage occurring in vaginal delivery, reducing retrograde menstruation132.

This association has led researchers to consider the role of surgical trauma and wound healing in lesion formation. It has been proposed that newly formed surgical scars may provide a favorable microenvironment for endometriotic lesion development, facilitating cell adhesion and proliferation through extracellular matrix remodeling, angiogenesis, and the release of growth factors133. Incisional endometriosis (i.e., endometriosis appearing in the cesarean section, episiotomy, or trocar site scars) is a well-documented phenomenon supporting this hypothesis52,134,135,136. These areas of surgical trauma could be potential sites where endometrial cells, either spread during surgery or regurgitated in retrograde menstruation, could implant and develop into new lesions109,128,135.

Coincidentally, perioperative immune suppression, triggered by surgical stress, may further impair the body’s ability to clear minimal residual disease and disseminated endometrial tissue109. Together, transient immunosuppression and local trauma may synergistically create a permissive environment for lesion establishment and recurrence137.

De novo lesion formation as a potential contributor to late endometriosis recurrence

Surgical factors alone cannot fully explain why some patients experience delayed or multiple recurrences despite apparently complete excision. Recurrence rates remain high over time despite advances in surgical techniques138,139,140, raising the possibility that some recurrent lesions may not stem from residual disease, but rather form de novo. Since surgery does not address the underlying pathophysiology of endometriosis, it is plausible that the disease reappears through the same mechanisms that led to its initial development, representing a progression of the natural course of the disease in susceptible individuals, rather than merely a consequence of incomplete treatment.

Several observations support the hypothesis of de novo lesion formation. First, recurrent lesions can appear in previously unaffected anatomical sites, including the contralateral ovary or other pelvic locations18,23,120. Second, recurrence rates tend to increase with longer follow-up periods25,44,120. This long-term progression has been interpreted by some authors as supporting de novo lesion formation, hypothesizing that endometriosis may eventually recur in all operated patients128,141. However, there is currently no consensus on whether reported timeframes are enough for the full development of new lesions and symptoms11,39. Lastly, the association between younger age and increased recurrence risk also supports this view, as it implies a longer time period for lesions to reappear before menopause38.

Assuming that endometriotic lesion recurrence not necessarily stems from residual disease shifts the focus towards the biological processes implicated in lesion initiation. Sampson’s theory of retrograde menstruation remains the most widely accepted explanation for the development of endometriotic lesions142 and is compatible with de novo recurrence. Mechanisms proposed to explain primary disease in this context, such as immune dysregulation143, stem cell involvement144, or pro-angiogenic, immunotolerant peritoneal niche145,146, may also be relevant in the context of recurrence. In this setting, implantation of regurgitated menstrual debris in the peritoneal cavity may be further facilitated by the permissive local environment created by tissue trauma and transient immunosuppression, as discussed above. From this perspective, late recurrence may reflect an individual’s biological predisposition to lesion formation.

Future directions: recurrence as a conceptual window into endometriosis pathophysiology

Current evidence indicates that endometriosis recurrence is a result of multiple, overlapping processes, including residual disease, surgery-related effects, and de novo lesion formation, whose relative contributions may vary depending on each patient’s surgical context and baseline biology. Understanding the biological basis of disease recurrence is essential to move beyond its clinical characterization and towards identifying the core processes sustaining endometriosis. To date, however, most research has focused on surgical determinants, leaving these underlying susceptibilities largely unexplored.

It is plausible that women prone to recurrence may harbor intrinsic biological traits that favor lesion re-establishment, such as impaired immune clearance of menstrual debris143, epigenetic alterations that enhance cell survival and adhesion147, increased presence of endometrial stem cells144, and a peritoneal microenvironment more permissive to implantation145,146. These mechanisms, proposed as contributors to disease onset, may similarly promote lesion reappearance in biologically susceptible individuals, a hypothesis that warrants future investigation.

Hence, viewed through this lens, recurrence becomes not just a clinically relevant outcome, but also a conceptual and biological model for investigating the core processes of endometriosis. Patients who develop recurrence despite apparently complete excision may represent an informative subgroup in whom disease drivers remain active or re-emerge after surgery. Their disease course could provide insights into key drivers of lesion initiation, including microenvironmental, immunological, or molecular alterations that facilitate the growth of residual cells or the implantation of disseminated or regurgitated endometrial cells. Understanding these pathways could help clarify whether recurrence is more a reflection of disease progression than surgical failure, and whether some patients exhibit a more aggressive disease phenotype.

A deeper understanding of the biological basis of recurrence could also inform prevention and treatment strategies. If early postoperative or baseline characteristics could predict recurrence, they might guide surgical decisions, follow-up intensity, and duration of hormonal suppression. Moreover, identifying mechanistically relevant molecular signatures may provide not only predictive tools but also functional biomarkers, enabling patient stratification, precise monitoring of disease progression, and the development of targeted, mechanism-based therapies, potentially beyond hormonal treatment.

In this light, recurrence of endometriosis lesions should be reframed not only as a clinical challenge, but also as a conceptual opportunity to investigate the pathogenic mechanisms of endometriosis: a natural stress test that reveals the processes most essential to lesion survival and reappearance, opening a window into the pathogenesis of the disease itself.

Conclusions

Recurrence remains a major challenge in the clinical management of endometriosis and a source of uncertainty for both patients and professionals. Conceptual and methodological differences across studies complicate the accurate interpretation of clinical outcomes. The adoption of a standardized, multidimensional definition of recurrence, together with uniform outcome measures systematically collected during long follow-ups, will be crucial to improve research comparability. These efforts should ideally build on and expand the existing Endometriosis Phenome and Biobanking Harmonisation Project (EPHect) surgical standard operating procedures148 to incorporate longitudinal outcome reporting and facilitate multicentric studies on endometriosis recurrence.

Beyond methodological improvement, endometriosis recurrence represents a biologically informative event and offers a unique opportunity to uncover key pathways involved in lesion initiation and survival. Integrating clinical and molecular data from well-annotated longitudinal cohorts could drive the development of predictive tools, mechanistically relevant biomarkers, and novel therapeutic targets, not only laying the foundation for patient risk stratification and personalized long-term management but also improving our broader understanding of endometriosis pathogenesis.