The surgical removal of endometriosis lesions is a method of treatment with proven effectiveness in reducing pain and improving fertility in patients with endometriosis. Although one of the most effective forms of treating pain in endometriosis is hormonal treatment, it cannot be used in patients trying to become pregnant because it does not improve the effect of efforts at any stage of treatment [1]. A thorough discussion of endocrine therapy is beyond the scope of this manuscript, which mainly focuses on endometriosis in the context of fertility disorders. It has been known for a long time that the destruction of lesions leads to the alleviation of immune disorders. The removal of the ectopic foci, which release immunosuppressive agents and impair the ability of NK cells to eliminate endometrial cells from the peritoneal cavity, reduces their negative impact [45]. Thus, the complete eradication of endometriotic lesions should be considered as the most effective form of causative treatment that also affects the immune system and offers high potential for improving the pregnancy rate. The probability of spontaneous conception is represented by the endometriosis fertility index (EFI) that is calculated in post-operative patients. Up to 74.9% ± 4.2% of patients with the best prognoses conceive spontaneously within three years following the radical removal of lesions. The stage of endometriosis is of lesser importance in such cases. Factors that are taken into account include age, how long the patient has been trying to conceive, any previous pregnancies, and the condition of the reproductive organs after the endometriosis surgery [46].

Except for observational, cross-sectional studies, it is difficult to perform reliable investigations on endometriosis patients. The majority of immunological endometriosis research is performed on animals. In nature, spontaneous endometriosis affects only those mammalian species that menstruate, including: primates, some bat species, the spiny mouse and elephant shrew [47]. Many animal models are not fully reliable, since endometriosis is induced artificially and does not represent all the phenomena present in the disorder. Generally, with various technical modifications, endometrial tissue (sometimes along with fragments of the myometrium) originating from the uterine cavity is translocated into ectopic locations. The development of this tissue is then intended to represent the various processes that occur during the development of endometriosis. The entire model is also disturbed by the necessity to undergo procedure, which may also lead to significant injury and inflammatory or infectious stress. Other models that must utilize primates are expensive and troublesome for many reasons. Moreover, artificial endometriosis behaves differently than spontaneous endometriosis, even in the same experimental animal model or in the same animal [48,49].

In order to verify the influence of the immune system and the possible role of immunosuppression on endometriosis, an animal model was implemented despite the above reservations. Olive baboons develop endometriosis spontaneously; thus, they are a better model for tracking the course of the disease. As early as in 1995, D’ Hooghe treated female olive baboons with immunosuppressive therapy combined with methylprednisolone and azathioprine—similar doses to the pharmacotherapy after an allogeneic heart transplant. Female baboons with spontaneous endometriosis showed an increase in the size and surface area of the lesions and a more advanced stage of the disease according to the American Fertility Society (AFS) classification. Newly developed ovarian endometrioid cysts were the result of azathioprine-only treatment. After immunosuppression discontinuation, at least a partial remission of the lesions was observed. At the beginning of the experiment, healthy baboons were subjected to the therapy and remained endometriosis-free. Moreover, surgically induced lesions failed to respond to immunosuppression [50]. This study was a significant warning and indicated that the indiscriminate and arbitrary use of immunological treatment, previously eagerly proposed to patients with infertility or habitual miscarriages, may abolish immune surveillance, paradoxically aggravate the course of endometriosis and worsen the effects of infertility treatment.

Also worth mentioning are the outcomes on short-term immunotherapy using glucocorticosteroids (GCS), derived from the study applying the method of in vitro fertilization (IVF) in patients with endometriosis and patients with a tubal factor of infertility. In patients with endometriosis, the presence of autoantibodies was significantly more prevalent than in healthy controls [18,19] and patients with tubal factors only. Researchers have proven that transient prednisolone administration from the 3rd day of the menstrual cycle significantly raised the pregnancy rate in the endometriosis group with antibodies. After the establishment of pregnancy, steroid therapy was not prolonged and endometriosis was not found to further influence the course of the pregnancy [51]. Another study showed that steroid use during an IVF cycle in autoantibody positive patients was associated with a higher clinical pregnancy rate (80% vs. 0%, p < 0.05); however, the effect of steroids in autoantibody-negative patients was not beneficial (46.7% vs. 45%, p = NS) [20].

A positive effect of drugs inhibiting the action of TNF on the development of endometriosis has been shown in animal models. There was a reduction in lesion size and an improvement in the fertility indicators, such as apoptosis and embryotoxicity. However, the administration of the TNF antagonist, imfliximab, in women with endometriosis did not reduce the pain symptoms or the size of the lesions. Its positive effect on fertility has also not been analyzed. Etanercept, another TNF antagonist, has been shown to have a positive effect on pregnancy rates in patients with endometriosis-associated infertility or ovarian endometrial cysts undergoing IVF treatment [52,53]. The data on the use of etanercept in other diseases indicate its high safety in early pregnancy. Unfortunately, the available data are scarce and do not allow this type of treatment to be recommended for routine usage at the moment [10].

Another postulated treatment proposition involves the Bacillus Calmette–Guérin (BCG) tuberculosis vaccine—one of the most effective immunotherapeutic agents with long application experience. It has been successfully used in infectious diseases and oncology for a long time and may find yet another application in endometriosis therapy. BCG administration triggers NK cells and induces factor recruitment, increasing the cytotoxicity of the NK cells. Animal studies have confirmed that subjects previously treated with the BCG vaccine had significantly endometriosis lesions [54]. BCG vaccination alters the secreted pro-inflammatory cytokines, IFN-γ, TNF, IL-2, IL-1a, IL-17 and IL-6; the Th2 cytokines, IL-4, IL-5, IL-13, and IL-10; and the chemokines, IP-10, MIP-1a, and IL-8, as well as the growth factors, granulocyte macrophage colony-stimulating factor (GM-CSF) and granulocyte colony-stimulating factor (G-CSF). Moreover, BCG upregulates the peripheral population and the function of NK cells, and also reprograms monocytes. As mentioned above, a polarization towards a Th2 response has been observed in endometriosis [23]. BCG is a strong inducer of Th1-type immunity and has been reported to protect against Th2-driven disorders. Despite promising opportunities, there is still a lack of data to demonstrate the effectiveness of this strategy in the human model [55]. There have also been attempts to develop and apply more specific vaccines with an immunotherapeutic effect in endometriosis. Studies with the use of a vaccine imitating the antigens of endometriosis and uterine fibroids seem to be interesting. In an animal model, a significant impact of administering this type of experimental vaccine on the subsequent possibility of induction of endometriotic foci has been demonstrated. In these immune-protected rats, the probability of the induction of endometriosis decreased from 69.6% to only 4.3% [56]. The administration of the vaccine before the induction of endometriosis prevented the increase of Th-2 dependent cytokines: IL4, IL6, and IL-10, which is typically observed during the development of endometriosis [57].

The central role of NK cells in the immune dysregulation exerted by endometriotic implants, as well as the rapid normalization of their activity after endometriosis removal, is discussed above [45]. Therefore, the possible application of immunotherapy to influence NK cell activity has also been suggested. NK cells have inhibitor receptors on their surface that suppress their activity against ectopic or malignant cells. Among receptors, PD1 that binds to the PDL1 ligand has already successfully been implemented in cancer immunotherapy. This kind of treatment may attempt to influence endometrial cell rescue, abolish the suppression of the supervisory function of NK cells, and allow for the removal of ectopic endometrial cells. The suggested immunotherapy postulates the application of medicines already used to treat other conditions, such as cancers, to facilitate the detection and elimination of endometrial cells through apoptosis [58].

Another potentially helpful agent in endometriosis treatment is low-dose naltrexone (LDN). As an opioid receptor antagonist, it has long been used in addiction therapy, chronic pain syndromes, and autoimmune diseases. The administration of low doses, as opposed to high doses, stimulates endorphin release and allows for immune system modulation by the central nervous system. Among other effects, LDN modulates the function of immune cells, such as bone marrow dendritic cells and macrophages, by methionine enkephalin (OGF), a potent antitumor agent, as well as by the inhibition of Tool-like receptor 4 (TLR-4) [59]. Moreover, LDN represses the proliferation of CD4+ and CD8+T cells and B lymphocytes in the spleen and lymph nodes [60]. The drug exerts slight transient side effects in 5% of patients, including sleep disturbances, daydreaming, nausea, headaches, and mucosal dryness. Naltrexone increases the efficacy of ovulation stimulation by anti-estrogens—being a cheaper and safer application than gonadotropins. An analysis of naltrexone use in polycystic ovary syndrome revealed a statistically significant decrease in BMI and insulin, LH/FSH, and testosterone levels in patients treated with naltrexone, which resulted in a 33% pregnancy rate. In the majority of patients with pelvic pain syndrome, LDN therapy allowed at least a stabilization or reduction in symptoms without compromising fertility [61,62].

A common problem in endometriosis patients that could be an additional reason for infertility is recurrent luteinized unruptured follicle syndrome (LUF), which affects up to 30% of menstrual cycles in endometriosis patients, compared to 6% in healthy women. Particularly in patients with ovarian endometriomas, LUF may affect more than 50% of cycles [63]. The lack of proper ovulation in such cycles makes it impossible to become pregnant and if LUF occurs regularly, it may be the only cause of infertility in this group of patients with endometriosis. Interestingly, neutrophils appear to play an essential role in ovulation as they are required to release metalloproteinases and digest the sheath of the Graafian follicle. In the absence of proper stimulation by G-CSF, neutrophils cannot be appropriately recruited into the ovary, and the follicle cannot release the oocyte correctly. The significantly increased frequency of the LUF syndrome in endometriosis is explained by the dysregulation of the neutrophilic function resulting from an impaired immune response. A logical treatment for LUF seems to be the external administration of G-CSF. A study on a patient with recurrent LUF was conducted in which doses of 30–45 thousand units of G-CSF were administered 24–48 h before chorionic gonadotropin administration in a stimulated cycle. After each cycle during which G-CSF was incorporated into the stimulation protocol, regular ovulation occurred and the post-ovulatory phase was stable and unimpaired. Where no G-CSF was administered, ovulation was abnormal with functional cyst (LUF) formation. Up to 90% of patients with infertility ovulated after the administration of the stimulating factor, compared to only 50% of the controls [64]. Therefore, G-CSF appears to be a safe and effective treatment for LUF syndrome associated with endometriosis.

Finally, it is worth pointing out a simple method of immunotherapy already used in practice, the detailed description and mechanisms of which go beyond the scope of this study. The immunotherapeutic effect in endometriosis-related infertility can be attributed to the intra-uterine administration of an ethiodized oil contrast agent, Lipiodol. A significant improvement in fertility, especially in patients with endometriosis, was noticed after the application of this agent, which was used primarily to assess the patency of the Fallopian tubes. The increase in fertility is estimated at 2 to 5 times within 6 months. Finally, in the three-year follow-up, more than half of the patients became pregnant, regardless of whether they were burdened with endometriosis or with so-called idiopathic infertility, which can also be associated with an undiagnosed immune background. This initially unintended effect was noticed after an analysis of the unexpectedly high pregnancy rates after patency examination if oil contrast was used. It was proven that Lipiodol upregulates uNK and dendritic and Treg cells in the endometrium, as well as in the peritoneal cavity. The application of an oil contrast also inhibits peritoneal lymphocyte and macrophage function. At the moment, however, it is not clear which mechanism plays the greatest role in improving fertility. Currently, oil flushing is becoming more and more widely used as an independent method of infertility treatment in endometriosis and idiopathic infertility [65,66,67,68].