Overall, the cumulative lifetime risk of being diagnosed with CRC, EC, and OC in the general population is 4.3%, 3.1%, and 1.3%, respectively. In the subpopulation affected by LS, the cumulative risk ranges from 9% to 61% for CRC, from 19% to 71% for EC, and from 3% to 14% for OC
[6][7][8][9][10][11]. Notably, EC and OC are the second and third most common LS-related malignancies, respectively, after CRC
[12]. However, after stratification for LS genetic variants, only
MLH1 mutant carriers (MLH1mut) have a higher incidence of CRC than EC, whereas
MSH6 and
PMS2 mutant carriers (MSH6mut, PMS2mut) have a higher incidence of EC followed by CRC and OC, and for
MSH2 mutant carriers (MSH2mut) the incidences of CRC and EC are superimposable
[13].
2. Lynch Syndrome-Associated Gynecologic Tumors
2.1. Lynch Syndrome-Associated Endometrial Cancer
Endometrial cancer is the second most common malignancy of the female genital tract worldwide, and it is the first most common gynecologic tumor in Western countries, with globally increasing incidence and mortality
[15][16].
The most common risk factors for endometrial neoplasia include unopposed hyperestrogenism conditions (associated with diabetes, obesity, early age at menarche and late age at menopause, nulliparity, high postmenopausal estrogen concentrations), older age, history of breast cancer, and long-term use of tamoxifen
[17]. However, approximately 5% of endometrial neoplasms are associated with a genetic predisposition, and 2–3% of all ECs are caused by a germline mutation in one of the MMR genes (
MLH1,
PMS2,
MSH2 and
MSH6)
[18][19]. However, the incidence in selected populations at higher risk for LS (based on family, personal, and pathologic criteria), evaluated by NGS, has been reported to be approximately 13%
[20].
LS-related EC is typically characterized by an endometrioid histotype, lower uterine involvement, peritumoral and tumor infiltrating lymphocytes (≥40 TIL/10HPFs), with more CD8+, CD45RO+, and PD1+ T cells at the invasive margin compared to sporadic MMRd EC, morpho-phenotypic heterogeneity, higher pathologic grade or undifferentiated forms, frequent lymphovascular invasion, deeper myometrial invasion, in particular with a typical MELF pattern
[21], frequent association with synchronous ovarian cancer, early age at diagnosis (45–55 years), and lower body mass index
[22].
In general, the MMRd group represents the 20–30% of all ECs and is considered a molecular subset with a good-to-intermediate prognosis, regardless of the histotype
[23]. Among MMRd patients, approximately 10–30% of ECs were observed to be associated with a germline variant
[18][24]. More specifically, methylated MMRd ECs seem to have a worse prognosis compared to mutated MMRd ECs, allowing a substratification of the MMRd/MSI group
[25]. However, although histopathologic features can significantly improve the efficacy of MMRd/MSI detection and the identification of those sentinel cases highly suspicious for LS, other authors have demonstrated that 58% of LS-related ECs do not exhibit the classically described morphologic MSI tumor features
[26]. Therefore, the use of Universal Screening to support a morphologic suspicion of LS has recently been recommended in the latest NCCN guidelines
[13].
2.2. Lynch Syndrome-Associated Ovarian Cancer
Ovarian cancer is the second most common gynecologic tumor in developed countries, with an incidence of more than 19,000 new cases in the United States and a higher mortality rate compared to EC
[15]. More than one-fifth of ovarian tumors are caused by a hereditary susceptibility, and in approximately 65–85% of these cases, the genetic abnormality is a germline mutation in the BRCA genes. However, several other suppressor genes and oncogenes have been associated with hereditary OC, and LS is estimated to be the cause of 10–15% of hereditary OC
[19]. Approximately 80% of LS-related OCs are diagnosed at stage I or II, mostly non-serous, with most of them presenting with endometrioid histology, with improved survival rates
[27].
3. Universal Screening
The evaluation of MMRd and/or MSI in EC has been shown to be relevant for LS screening, histo-molecular diagnosis, prognostic risk stratification, and as a predictive biomarker of response to immunotherapy
[24].
Prior to the introduction of Universal Screening, the diagnostic management of LS was mainly based on Selective Screening, which included clinical criteria based on the evaluation of family and personal history and the clinicopathological features of the tumor (Amsterdam criteria 1990
[28][29], Bethesda criteria 1997
[30], and Society of Gynecologic Oncology 20–25% and 5–10% criteria 2007
[31]). However, these have been limited in their application to clinical practice due to their complexity and the frequent lack of complete family history data. In particular, Ryan et al. reported that these criteria were only able to identify the 36% (Amsterdam criteria 1990), 58% (Bethesda criteria 1997), 71% (Society of Gynecologic Oncology 20–25% criteria), and 93% (Society of Gynecologic Oncology 5–10% criteria) of patients who required further LS testing
[32]. Selective Screening has traditionally been applied only to a selected population of EC or CRC patients at higher risk of being LS carriers based on family and/or personal history and pathologic features suspicious for MSI
[33]. In contrast, Universal Screening has been applied to all newly diagnosed EC and CRC cases; with studies comparing the two screening approaches showing a greater sensitivity of Universal Screening in detecting patients with LS
[34]. Therefore, a Universal Screening for LS diagnosis by the identification of MMRd in the tumor tissue of all newly diagnosed CRC and EC cases has recently been proposed
[35][36][37].
The first step of Universal Screening is performed by IHC, a technique capable of evaluating the somatic loss of expression of the four proteins (MLH1, MSH2, MSH6, and PMS2) encoded by the MMR genes on a tumor sample. The flowchart for the identification of LS cases through Universal Screening is represented in
Figure 1. Although, in some instances, a pathologist may encounter various analytical problems or diagnostic pitfalls, IHC is considered the standard, because it is more cost effective, has a shorter turnaround time, and is widely available. IHC can correlate with morphology and provide useful information about which of the four proteins is missing, thus guiding genetic testing. In addition, IHC is considered superior to MSI evaluation as a first-line testing strategy, due to the high frequency of microsatellite stable (MSS) ECs harboring mutations in MSH6, which can only be detected by the IHC approach in some cases
[38][39]. Finally, considering the high frequency of MMRd and MSI observed in the peritumoral endometrial background in LS-EC patients, the IHC evaluation of MMR protein expression in the benign endometrium could be further explored as a possible integrative test to the LS screening algorithm
[40]. As mentioned previously, the loss of MMR protein expression, which occurs in approximately 20–30% of patients with EC
[18], requires further evaluation, and genetic counseling is always particularly recommended. About 10–30% of MMRd/MSI ECs have been observed to be associated with a germline variant
[18][24], and the loss of expression of the protein encoded by the
MLH1 gene represents the majority of all MMRd cases. Most of these patients do not have a germline mutation, but a loss of expression is due to somatic biallelic hypermethylation of the
MLH1 promoter, as reported for CRC
[41], to somatic biallelic
MMR gene mutations, or to a somatic mutation associated with a loss of heterozygosity (LOH) of the other allele; alternatively, it is due to a secondary epigenetic silencing of
MSH6 after neoadjuvant radio-chemotherapy treatments. In EC cases characterized by defective MLH1 expression, a somatic ‘reflex test’ is required to assess the
MLH1 promoter methylation status. In contrast, the loss of expression of any of the proteins encoded by the
MSH2,
MSH6, or
PMS2 genes, and the absence of
MLH1 hypermethylation always requires a germline evaluation by NGS.
Figure 1. Flowchart for the identification of Lynch Syndrome by Universal Screening. IHC: immunohistochemistry; EC: endometrial cancer; CRC: colorectal cancer; MMRd: mismatch repair deficient; MMRp: mismatch repair proficient; LS: Lynch Syndrome.
4. Prevention of Lynch Syndrome-Associated Gynecologic Tumors
4.1. Surveillance and Surgical Prophylaxis
As mentioned in the previous sections, women with LS have a 19–71% and 3–14% lifetime risk of developing endometrial cancer (EC) and ovarian cancer (OC), respectively, depending on the specific mismatch repair gene mutated
[6][7][8][9][10][11]. The gene-specific cancer risk for EC and OC and the average age at diagnosis are summarized in
Figure 2.
Large retrospective population studies have reported a cumulative risk of developing EC by the age of 70 years in patients with LS to be approximately 34–54% for MLH1mut, 21–57% for MSH2mut, 16–49% for MSH6mut, and 24% for PMS2mut
[42][43]. Similar results were confirmed also in a large prospective multicenter study
[5].
There is less evidence in the literature for LS-associated OC compared to the more commonly found LS-associated EC. Engel et al. found similar cumulative odds of lifetime OC diagnosis for MSH2mut and MSH6mut and a reduced risk for MLH1mut
[44]. Similarly, previous data reported that MSH2mut had almost twice the OC incidence rate observed for MLH1mut
[45][46]. Specifically, the cumulative risk of developing OC by the age of 70 years was reported to be 4–20% for MLH1mut, 8–38% for MSH2mut, <1–13% for MSH6mut, and, similarly to the overall population, 1.3–3% for PMS2mut
[5][42][43].
The ESGO-ESTRO-ESP 2021 consensus recommends surveillance for EC in LS pa- tients starting at the age of 35 years by annual transvaginal ultrasound (TVUS) and annual or biennial endometrial biopsy; and prophylactic hysterectomy and bilateral salpingo- oophorectomy should be considered to prevent both EC and OC at the end of the childbear- ing age and preferably before the age of 40 years. In addition, estrogen replacement therapy should be recommended in premenopausal women undergoing prophylactic surgery with hysterectomy and bilateral salpingo-oophorectomy.
More recently, the NCCN 2022 guidelines have outlined a series of recommenda- tions for the preventive management of LS-related tumors based on the specific mutational pattern carried. Specifically, surveillance by endometrial biopsy should be performed annu- ally or biennially beginning at the age of 30–35 years; in addition, TVUS in postmenopausal women, although not considered sufficiently sensitive or specific, may be performed for screening at the clinician’s discretion in MLH1mut, MSH2mut, MSH6mut, and PMS2mut. Serum CA125 is an additional ovarian marker that could be incorporated into OC screening in any mutation status. Moreover, any abnormal uterine bleeding or postmenopausal bleed- ing, pelvic or abdominal pain, bloating, increased abdominal circumference, and any other abnormal and newly onset sign or symptom should be reported promptly to a clinician in any mutation-carrier patient. Regarding prophylactic surgery, a total hysterectomy and bilateral salpingo-oophorectomy are recommended in MLH1mut and MSH2mut to reduce the risk of both EC and OC. In these cases, the timing of surgery should be individualized based on whether childbearing is complete, comorbidities, family history, and LS gene. In MSH6mut and PMS2mut, a total hysterectomy should be recommended to reduce EC risk after completion of the reproductive desire; however, evidence is lacking to make a specific recommendation for risk-reducing salpingo-oophorectomy, and in particular the PMS2 pathogenic variant appears to have no greater than average risk for OC and may consider postponing surveillance and may reasonably elect not to have oophorectomy. Therefore, for both MSH6mut and PMS2mut, the decision on whether undergo bilateral salpingo- oophorectomy should be discussed with the patient and individualized accordingly.
Figure 2. Gene-specific cancer risk and average age at diagnosis
[5][42][43][47][48].
4.2. Chemoprevention
Studies in the general female population have reported that conditions increasing the estrogen bioavailability unopposed by progesterone (obesity, early age at menarche, late age at menopause, nulliparity, and the use of estrogen-only menopausal hormonal therapy) increase the risk of EC. Conversely, the use of hormonal contraceptive, higher number of pregnancies, and later age at first and last live birth have been shown to decrease the risk of developing gynecologic tumors
[49][50].
Although there are few data on how hormonal status might influence cancer risk in women with LS, some evidence reported that combined oral contraceptives or progestins-alone (oral, injectable, or intrauterine device, IUD) might decrease EC risk
[51]. Progestin-based chemoprevention appears to reduce EC risk in women regardless of MMR status
[51][52]. In addition, depo-medroxyprogesterone acetate or oral contraceptive combination have been observed to reduce endometrial proliferation in women with LS
[53]. Therefore, the Manchester International Consensus Group recommends that any MMR pathogenic variant carrier consider the option of oral contraception, whenever contraception is desired and after refusal of prophylactic surgery, in accordance with a clinician, to reduce the risk of both EC and OC
[36]. However, prospective data are needed before including oral contraceptives in the gynecologic cancer prevention management of any LS carrier.
Aspirin has a role in the prevention of LS-associated tumors. Studies on CRC found explanations for aspirin’s efficacy in cancer prevention by both downregulation of local and systemic inflammation and WNT/beta-catenin signaling pathway. In fact, aspirin inhibits prostaglandin endoperoxide synthase-mediated conversion of arachidonic acid to prostaglandins either through a direct effect on mucosal cells or through a paracrine effect mediated by platelet synthesis of thromboxane A2. Moreover, aspirin may also inhibit WNT signaling either directly or by downregulation of prostaglandin synthesis
[54]. The CAPP2 study demonstrated that daily aspirin at a dose of 600 mg may be able to reduce the incidence of CRC when compared with placebo; however, the authors did not find significant differences in the extraintestinal LS-related tumors subset
[55]. The CAPP3 trial (NCT02497820) is evaluating the role of different doses of aspirin (100, 300, and 600 mg) in the same patient population. Although not proven to be effective in LS-related gynecologic tumors, aspirin may be administrated to reduce the risk of cancer in LS carriers
[13]; however, the dose and duration of treatment have not been established.
4.3. Vaccination
LS-associated tumors are characterized by a prominent local immune response and generally exhibit a rich lymphocyte infiltration. This suggests that the immune system may play an active role in the surveillance and natural history of these tumors
[56]. MSI tumors accumulate numerous frameshift mutations. Frameshift mutations affecting cancer-related genes can promote tumorigenesis; therefore, such recurrent frameshift mutations may give rise to shared immunogenic frameshift peptides (FSPs) that represent a suitable vaccine against MSI-related cancers
[57]. Several studies in the literature have validated the strong immunogenicity of tumor-specific antigens derived from shared frameshift mutations in MSI-cancer and LS patients, suitable for the design of common “off-the-shelf” cancer vaccines
[58][59]. Gebert et al. identified four shared FSP neoantigens (Nacad [FSP-1], Maz [FSP-1], Senp6 [FSP-1], and Xirp1 [FSP-1]) capable of inducing CD4/CD8 T cell responses in mouse models, and demonstrated that combination of only four FSPs significantly increased FSP-specific adaptive immunity, reduced tumor burden, and improved survival in LS mice with CRC
[60].
In addition, the Phase Ib/II clinical trial NCT05078866 is currently evaluating the safety, tolerability, and efficacy of the Nous-209 cancer-preventive vaccine in LS carriers. Nous-209 vaccine is composed of an adenoviral tumor-specific neoantigen priming vaccine GAd-209-FSP (GAd20-209-FSPs) and MVA tumor-specific neoantigen boosting vaccine MVA-209-FSP (MVA-209-FSPs), and estimated study completion data are expected in July 2025.
Although results from pivotal clinical trials are still lacking, early evidence of the prophylactic role of vaccination in reducing both cancer risk and cancer progression in patients with LS are promising and warrant further investigation.
5. Management of Patients with LS-Associated Gynecologic Cancers
5.1. Surgery and Conservative Treatment Management
The management for LS-associated EC and OC is similar to that of their sporadic counterparts. According to international guidelines
[61], the standard treatment for EC consists of a total hysterectomy with bilateral salpingo-oophorectomy. However, in selected patients with atypical endometrial hyperplasia (EAH), endometrioid intraepithelial neoplasia (EIN), or grade 1 endometrioid EC without (or with minimal) myometrial invasion, medical management may be considered to preserve fertility or avoid potential surgical complications in obese women
[62]. Accepted medical treatments include the use of oral megestrol acetate (MA) or medroxyprogesterone acetate (MPA), and levonorgestrel-releasing intrauterine device (LIUD)
[61]. Several studies have been already conducted to analyze the response to conservative treatment and progression rate in MMRd patients. However, few data are available for the specific subset of patients with LS.
According to the available evidence, MMRd status and LS appear to be associated with a lower response rate to conservative treatment, a higher risk of recurrence, and worse prognosis compared to the MMRp subset.
5.2. Systemic Treatment
If adjuvant radiotherapy and/or systemic treatment is required, patients with LS are usually treated according to the standard of care as defined by international guidelines, which includes a platinum-based regimen
[24][61]. However, in the subset of LS-related cancers, additional and specific treatments can be considered for the second-line management.
Genomes of MMRd cancers harbor high microsatellite instability (MSI-H) and express multiple somatic mutations encoding potential neoantigens. Thus, these subgroups of tumors are likely to be immunogenic (“hot tumors”), triggering upregulation of immune checkpoint proteins.
Pembrolizumab, an anti-programmed death-1 monoclonal antibody, acts as an antitumor weapon against MSI-H/MMRd immunogenic tumor environment
[63][64]. In the KEYNOTE-158 trial, in a setting of previously treated advanced extraintestinal MSI-H/MMRd solid tumors (including both EC and OC), pembrolizumab demonstrated robust evidence of objective response rate (ORR) and durable benefit
[65]. As a result, current available guidelines recommend the addition of pembrolizumab in the setting of advanced/metastatic disease for second-line treatment of MSI/MMRd cancers
[24]. However, an investigator-initiated Phase II trial (NCT02899793) evaluating the safety and efficacy of pembrolizumab in patients with MMRd and/or MSI-H EC showed, in 24 evaluable patients, differences in terms of ORR, progression-free survival (PFS) and overall survival (OS) among Lynch/Lynch-like tumors (somatic MMR mutations) and MLH1-methylated patients. Specifically, 100% of Lynch/Lynch-like patients achieved a complete response (CR) or partial response (PR), while only 44% of MLH1-methylated patients achieved an ORR. Similarly, 3-year PFS and OS were improved in the first group (100% vs. 30% and 100% vs. 43%, respectively)
[66][67]. In addition, pembrolizumab was approved in combination with lenvatinib for the treatment of patients with relapsed EC after a failure of a platinum doublet regimen based on the results of the KEYNOTE-775 trial
[68]. This study compared the combination of pembrolizumab and lenvatinib with physician’s choice chemotherapy, and showed an advantage in PFS and OS, both in the overall population and in the MMRp subgroup (PFS 7.2 vs. 3.8 months and 6.6 vs. 3.8 months and OS 18.3 vs. 11.4 months vs. 17.4 vs. 12.0 months, respectively).
Promising results are also reported from the GARNET trial, in which two groups of advanced EC patients (one with MSI-H/MMRd EC and one with proficient/stable [MSS/MMRp] EC) received dostarlimab, a humanized monoclonal antibody that binds with high affinity to PD-1, showing an ORR of 43.5% in the first group, with a more acceptable toxicity profile compared to pembrolizumab
[69]. In this case, no differences in ORR based on MMR protein loss pattern or MMR gene methylation/mutation were reported, although the trial was not powered to evaluate this outcome
[70].
Ongoing trials are currently evaluating the potential benefit of immunotherapy, also in the setting of first-line treatment for MSI/MMRd tumors.