1. Sex and Cancer Immunity
1.1. The Immune System in Female Patients
The response to immunotherapy is influenced by many factors, some yet to be discovered, but the intrinsic characteristics of the tumor and of the tumor environment are key in provoking an immune response
[1]. Sex differences are also applicable in the immune system. In general, women have stronger innate and adaptive immune responses than men, which is illustrated by the higher incidence of autoimmune diseases in women
[2]. The immune system differs between males and females due to genetic differences and sex-specific levels of hormones (estradiol, progesterone, and androgens).
The X chromosome contains several immune-related genes with different inheritance patterns for women and men
[3]. Nearly all immune cells express receptors for sex hormones that may influence the expression of several immune-related genes via responsive elements in promotor sites
[4]. Progesterone, although dependent on its concentration, has anti-inflammatory effects. Androgens suppress immune cells. Estradiol improves cell-mediated and humoral immune response. Low estrogen levels tilt the T helper (Th) response towards Th1 differentiation, enhancing cellular immunity. High estrogen levels shift the equilibrium towards the Th2 phenotype
[1].
Currently approved immunotherapies in non-small cell lung cancer (NSCLC) are monoclonal antibodies against the PD-1 (nivolumab, pembrolizumab, and cemiplimab), the PD-L1 (atezolizumab and durvalumab), or CTLA-4 proteins (ipilimumab). PD-1 is a cell surface receptor present on pro B and T lymphocytes and plays a role in downregulating the immune response and self-tolerance when binding to its ligands PD-L1 and PD-L2. This aids in preventing autoimmunity, but cancer cells may upregulate PD-L1 to escape immune-mediated elimination. PD-1 expression is influenced by estrogen and prolactin, and therefore, is sex-dependent
[5]. PD-L1 expression on tumor cells, on the contrary, should be less sensitive to the hormonal surroundings of the host. CTLA-4 is another protein receptor that is upregulated in activated T cells and responsible for suppressing the activity of other T cells
[6].
1.2. Immune Checkpoint Inhibition in Female Patients
The landmark KEYNOTE-024 trial compared pembrolizumab to platinum-based chemotherapy in treatment-naive patients. With a hazard ratio (HR) of 0.95 (95% CI 0.56–1.62) in females and 0.54 (95% CI 0.36–0.79) in males, this trial did not show a significant survival benefit in female patients
[7]. The similar EMPOWER-Lung 1 trial compared cemiplimab to chemotherapy with identical findings: in females, the hazard ratio was insignificant at 1.11 (95% CI 0.42–2.59) (males: 0.50; 95% CI 0.36–0.69)
[8]. The KEYNOTE-042 trial allowed for the lower PD-L1 expression of ≥1%. Again, pembrolizumab was superior to chemotherapy in men, with an OS hazard ratio of 0.68 (95% CI 0.53–0.88), but not in women (HR 0.78; 95% CI 0.53–1.15; results for the PD-L ≥ 50% subgroup as well as for the entire population)
[9]. Finally, the IMPOWER-110 trial also included PD-L1-positive patients (≥1%), comparing atezolizumab to chemotherapy. In the TC3/IC3 subgroup, the hazard ratio for women was 0.69 (95% CI 0.34–1.39), in contrast to the male ratio of 0.57 (95% CI 0.35–0.93)
[10]. Median OS values according to sex were only reported in (the appendix of) this last trial: 23.1 months for males (gaining a median of 10.0 months) and 17.8 months for females (gaining a median of 3.7 months). These findings were observed by a recent meta-analysis
[11].
In addition to this, the CHECKMATE-227 trial investigated an immunotherapy combination of nivolumab plus ipilimumab in PD-L1-positive patients (≥1%), with the chemo-free arm outperforming chemotherapy in males (HR 0.75; 95% CI 0.61–0.93), but not in females (HR 0.91; 95% CI 0.69–1.21)
[12]. It appears that women, unlike men, do not benefit (as much) from immunotherapy, whether in monotherapy or combined. In addition to fitness for chemotherapy and the need for a rapid response, biological sex might also be a criterion to consider when selecting first-line therapy in PD-L1-high NSCLC.
In patients with moderate (1–49%) and low (<1%) PD-L1 expression, immunotherapy combined with chemotherapy is the current standard of care. Here, the results of female and male subgroups are different. The two pembrolizumab trials showed an OS benefit in both males and females—females clearly more with non-squamous and males slightly more with squamous histology
[13][14]. A second meta-analysis of immunotherapy, alone or in combination with chemotherapy, confirmed that women derived a higher benefit compared to men from the combination of pembrolizumab/chemotherapy versus chemotherapy or any other treatment option (all P
interaction < 0.02)
[15].
In the IMPOWER-130 and -131 trials, females had a greater benefit in both histologies
[16][17]. In contrast, in the IMPOWER-150 trial, which added bevacizumab to the chemotherapy backbone, no real benefit was found in female patients
[18]. Finally, in the CHECKMATE-9LA trial, no clear difference in HR was observed
[19]. The proportion of female patients in all these trials was very heterogeneous: from 12% in the EMPOWER-Lung 1 trial to >40% in the IMPOWER-130 and KEYNOTE-024 trials, and statistical considerations of subgroup analysis may apply.
The fact that women are at an advantage with the addition of chemotherapy, as opposed to treatment with single-agent immunotherapy, could be the result of a greater mutational burden and tumor antigenicity in men. The recent EMPOWER-Lung 1 trial demonstrated that even within the high PD-L1 category, there were differences in response according to the PD-L1 level
[8], but there was no clear evidence that the average PD-L1 expression was lower in females
[20].
Considering the adverse events of immunotherapy, the female sex has been reported to be associated with greater toxicity of checkpoint inhibitors when inhibiting both CTLA-4 and PD-1/PD-L1
[21][22][23].
2. EGFR and ALK Inhibition in Females
Cancer in never-smokers is more common in females; in general, adenocarcinoma histology (NSCLC) is detected
[24]. Often, an oncogenic driver is identified in these patients. Mutations in the epithelial growth factor receptor (EGFR) are the most common driver in never-smokers and are more often found in women (odds ratio of 2.7; 95% CI 2.5–2.9)
[25], in addition to Asians and Caucasians and those with adenocarcinoma histology
[26]. For this reason, women are more often treated with targeted therapies and seem to benefit more compared to men in EGFR-mutated lung cancer. This is not the case in lung cancer with an anaplastic lymphoma kinase (ALK) fusion, where survival data are comparable between men and women
[27].
3. Sex and Lung Cancer Screening
3.1. Benefits from Lung Cancer Screening (LCS) in Female Patients
Although LCS trials were not specifically designed for women, LCS trials have revealed differences in lung cancer-specific mortality, LCS being far more beneficial in women than in men. Three randomized lung cancer screening trials have stratified the outcome data by gender: NLST, LUSI, and NELSON.
In the North American National Lung Screening Trial (NLST), the rate ratio for mortality from lung cancer among female participants in the low-dose computed tomography (LDCT) group, as compared to those in the chest-radiography group, was 0.80 (95% CI, 0.66 to 0.96) for a follow-up period of 12.3 years
[28]. The German Lung Cancer Screening Intervention Trial (LUSI) showed a significant benefit with respect to lung cancer mortality in the small subgroup of women who were invited to undergo screening (HR 0.31, 95% CI, 0.10 to 0.96)
[29]. The final publication of the Dutch–Belgian Lung Cancer Screening Trial (NELSON) mainly focused on the results of men due to the low number of women involved in the trial. Although not statistically significant, the data on the small subset of women showed more favorable effects of LCS for women than for men, with a rate ratio for death from lung cancer of 0.67 (95% CI, 0.38 to 1.14) at 10 years of follow-up. The magnitude of lung cancer-specific mortality reduction in women was even greater at 7, 8, and 9 years from baseline
[30].
A meta-analysis by Hoffman et al.
[31] revealed that women benefitted substantially more from screening, with a 31% relative risk reduction in lung cancer mortality compared to 14% in men. Risk reductions were statistically not significant (
p = 0.11). As previously mentioned, women were underrepresented in these LCS trials, so analyses for women were likely underpowered.
The inclusion criteria for LCS take into account the number of pack-years reflecting the severity of smoking history. In general, women accumulate fewer pack-years than men, resulting in differences in eligibility for LCS. Modeling studies have shown that expanding eligibility to include ever-smokers with less than 30 pack-years of exposure (20–29 pack-years) would not only increase the proportion of lung cancer deaths prevented by screening but would also reduce disparities in eligibility by sex
[32].
3.2. Harms of Lung Cancer Screening in Female Patients
Using low-dose CT, LCS is performed at radiation doses much lower than doses used in clinical practice for diagnostic chest CT imaging. Despite the lower radiation dose, radiation remains one of the harms associated with LCS. Researcher's knowledge from long-term studies on the impact of these levels of radiation exposure on cancer risk is very limited.
There is an interaction between radiation and smoking, with cancer risk from radiation generally being higher in the target population of smokers and former smokers. In addition to smoking and age, sex plays a role in the estimated risks of lung cancer associated with radiation. Excess relative risk differs between men and women and is higher in women than in men. Brenner et al. calculated the risks for yearly LDCT lung cancer screening: yearly screening would result in a 5% increase in the risk of lung cancer in women. In men, this increased risk would only be 1.5%
[33]. The precise mechanisms underlying the sex differences in radiation-induced cancers remain unknown. The roles of hormonal regulation, genetic risks, and X-linked factors still need to be determined
[34][35].
Rampinelli et al. retrospectively investigated the cumulative radiation exposure and lifetime attributable risk of cancer incidence associated with LDCT from a 10-year lung cancer screening program. They showed that the lifetime attributable risk of lung cancer was estimated to be about four times greater for women aged 50–54 years than for men aged 65 and older. The risk for other major cancers is up to three times greater. Both the increased radiosensitivity of women and the risk of breast cancer associated with chest imaging are postulated to be the cause of this difference
[36].
Overdiagnosis is another harm in LCS. Cancer overdiagnosis is the detection of asymptomatic cancers that would never have caused medical problems or harm during the patient’s lifespan because of death from other causes
[37]. In the era of LCS, it is defined as screen-detected cancer that would not have become symptomatic during a person’s lifetime. Blom et al. estimated overdiagnosis in lung cancer screening using the cumulative excess-incidence approach. With this approach, the difference in cumulative incidence between a screened group and a matched control group is attributed to overdiagnosis. Overall, the percentage of overdiagnosis of screen-detected cancers was higher in women (ranging from 5.7% in the 1990 cohort to 11.2% in the 1950 cohort) than in men (ranging from 61% in the 1990 cohort to 9.8% in the 1950 cohort) in all cohorts except the 1990 cohort. An explanation of this overdiagnosis may be related to the predominant slower-growing adenocarcinoma histology in women. The longer the preclinical duration of the disease, the higher the likelihood of overdiagnosis
[38].
3.3. Eligibility and Uptake of Lung Cancer Screening in Female Patients
Lung cancer screening targets high-risk participants, with smoking history being the most important risk factor. Smoking habits vary between sexes, with sex differences also varying between countries
[39]. Differences in current smoking habits will impact trends in lung cancer incidence in the upcoming decades. Currently, most screening programs are ‘one-size-fits-all’, with no different eligibility criteria for women and men. The risk of disease depends, however, on many individual factors, including sex and age.
A comparative simulation modeling study investigating seven selected risk factor-based screening scenarios showed a lower percentage of eligibility for women in all scenarios. In contrast, for all except one scenario, the percentage of mortality reduction was higher in women. The number of people needed to (ever) screen to prevent one lung cancer death (NNS) was lower for women compared to men for all scenarios. Due to radiation sensitivity in women (as discussed previously), the number of radiation-related lung cancer deaths was higher in women than in men for all seven scenarios
[32]. Lung cancer tends to be diagnosed in women at a younger age, an aspect that is not taken into account in the selection criteria for LCS
[40].