The Clinical and Molecular Profile of Lung Cancer Patients Harboring the TP53 R337H Germline Variant in a Brazilian Cancer Center: The Possible Mechanism of Carcinogenesis

In southern and southeastern Brazil, the TP53 founder variant c.1010G>A (R337H) has been previously documented with a prevalence of 0.3% within the general population and linked to a heightened incidence of lung adenocarcinomas (LUADs). In the present investigation, we cover clinical and molecular characterizations of lung cancer patients from the Brazilian Li-Fraumeni Syndrome Study (BLISS) database. Among the 175 diagnosed malignant neoplasms, 28 (16%) were classified as LUADs, predominantly occurring in females (68%), aged above 50 years, and never-smokers (78.6%). Significantly, LUADs manifested as the initial clinical presentation of Li-Fraumeni Syndrome in 78.6% of cases. Molecular profiling was available for 20 patients, with 14 (70%) revealing EGFR family alterations. In total, 23 alterations in cancer driver genes were identified, comprising 7 actionable mutations and 4 linked to resistance against systemic treatments. In conclusion, the carriers of TP53 R337H demonstrate a predisposition to LUAD development. Furthermore, our results indicate that environmental pollution potentially impacts the carcinogenesis of lung tumors in the carriers of TP53 R337H.


Introduction
In 2022, the American Cancer Society estimated a total of 236,740 new cases of lung cancer (LC), resulting in 130,180 deaths, accounting for about 25% of all cancer-related deaths [1].In Brazil, 32,560 new cases were predicted by 2023, and it is the fourth most common malignancy by incidence [2].Although most cases are sporadic, some cases are caused by external agents such as tobacco smoking, air pollution, and other environmental carcinogens [3].In the past few years, there has been a growing concern regarding outdoor air pollution as a carcinogen for LC, and its effect may be measured by particulate matter (PM) [4].The PM 2.5 (fine particulate matter with <2.5 µm of diameter) may enter the terminal bronchioles and alveoli, triggering local inflammatory processes [5].Incidence and mortality from LC with PM 2.5 levels have been demonstrated in North America, Asia, and Europe regardless of cigarette smoke exposure [6].
Germline pathogenic variants in the genes involved in carcinogen clearance, DNA repair, and cell cycle control may confer higher tumor-specific risk [7][8][9].Recently, a retrospective analysis of 7788 lung cancer patients who performed exome sequencing revealed the presence of germline pathogenic variant in 15% of cases, a higher prevalence than previously described for BRCA2, CHEK2, ATM, TP53, BRCA1, and EGFR [10,11].Notably, some of these genes are potentially targetable [11].
Li-Fraumeni syndrome (LFS) is a rare heritable condition resulting from germline pathogenic variants in the tumor suppressor gene TP53.Carriers are at a high lifetime risk of developing a large spectrum of cancers, including early-onset breast, sarcoma, central nervous system, adrenocortical, and lung malignancies [12].According to the IARC database, 3% of individuals harboring TP53 mutations develop LC, which represents the eighth tumor in incidence in the context of LFS [12].
In Brazil, the c.1010G>A p.Arg337His (R337H) variant in TP53 is present in 0.3% of the southern and southeastern population due to a founder effect [13].Unlike the vast majority of TP53 pathogenic variants, which occur in the DNA-binding domain (exons 5-8), R337H occurs at the oligomerization domain in exon 10 and has a milder penetrance than other variants.Previous studies in nonselected lung adenocarcinoma (LUAD) populations from Brazil found an enrichment of the R337H variant, with a prevalence ranging from 2.3% to 5.4% and a statistical correlation to HER family alterations [14][15][16].However, limited clinical and demographic information did not permit further conclusions.
The present study aims to describe the clinical and molecular features of R337H carriers diagnosed with LC in Brazil.Furthermore, we postulate some possible mechanisms involved in LC carcinogenesis in the context of R337H LFS.

Clinical Outcomes
Among those patients in early and locally advanced stages (stages I-III), 16 underwent surgery and attained a median disease-free survival (time from tumor resection to recurrence or death for any cause) of 106.9 months (range: 0.03-229.8months).Among them, 68.8% (n = 11) exhibited no evidence of disease, while 32.1% (n = 5) experienced recurrences (depicted in Figure 3a).At the latest follow-up, the median overall survival (time from initial cancer diagnosis to death from any cause) was not reached since none of the patients died (Figure 3b).Most cases indicated mutations in driver genes.A total of 23 alterations, comprising 7 actionable and 4 related to systemic treatment resistance, were described.In horizontal lines, individual characteristics of the patients are listed, such as average levels of exposure to PM2.5, gender, smoking, and evaluated genes.Each column indicates an individual patient.On the right, the genetic alterations are described and illustrated with colored squares.Abbreviations: del-deletion; indel-insertion or deletion; inser-insertion; PM2.5-particulate matter < 2.5mm of diameter.

Clinical Outcomes
Among those patients in early and locally advanced stages (stages I-III), 16 underwent surgery and attained a median disease-free survival (time from tumor resection to recurrence or death for any cause) of 106.9 months (range: 0.03-229.8months).Among them, 68.8% (n = 11) exhibited no evidence of disease, while 32.1% (n = 5) experienced recurrences (depicted in Figure 3a).At the latest follow-up, the median overall survival (time from initial cancer diagnosis to death from any cause) was not reached since none of the patients died (Figure 3b).
For individuals with advanced disease and EGFR mutations (n = 11), the median progression-free survival was 19.6 and 17.4 months for the first-generation (erlotinib and gefitinib) and third-generation (osimertinib) EGFR inhibitors, respectively, as illustrated in Figure 4a,b.In this context, the median overall survival was not reached, as shown in Figure 4c,d.Notably, one death occurred during osimertinib treatment, coinciding with the emergence of BRAF amplification.
In a patient harboring a somatic EGFR exon 20 insertion (D770_N771insY), an experimental EGFR oral inhibitor targeting exon 20 alterations was administered as the sixth line of treatment, resulting in partial response (for lesions in the lungs and brain) and symptom control for seven months (Figure 5a), followed by multifocal progression.Another patient with HER2 exon 20 insertion and liver metastasis received four lines of therapy (afatinib, carboplatin-paclitaxel, pertuzumab-trastuzumab-docetaxel, and trastuzumab-entansine, in that order) for a cumulative duration of 25 months until disease progression to the central nervous system (Figure 5b).
In the case of four metastatic patients who lacked EGFR family activation mutations and were subjected to chemoimmunotherapy, it should be noted that three of them     In a patient harboring a somatic EGFR exon 20 insertion (D770_N771insY), an experimental EGFR oral inhibitor targeting exon 20 alterations was administered as the sixth line of treatment, resulting in partial response (for lesions in the lungs and brain) and symptom control for seven months (Figure 5a), followed by multifocal progression.Another patient with HER2 exon 20 insertion and liver metastasis received four lines of therapy (afatinib, carboplatin-paclitaxel, pertuzumab-trastuzumab-docetaxel, and trastuzumab-entansine, in that order) for a cumulative duration of 25 months until disease progression to the central nervous system (Figure 5b).
In the case of four metastatic patients who lacked EGFR family activation mutations and were subjected to chemoimmunotherapy, it should be noted that three of them experienced an early loss of follow-up, occurring within a period of less than four months, rendering them ineligible for inclusion in the clinical outcomes analysis.The remaining patient died after 42 months of systemic treatment.

PM 2.5 Measurements
All cases were categorized across 17 cities in six Brazilian States (as outlined in Table 3), with the majority (n = 8) situated in the state of São Paulo, followed by Minas Gerais (n = 3), Paraná (n = 3), Federal District (n = 1), Goiás (n = 1), and Santa Catarina (n = 1).These cities correspond to three distinct regions within Brazil, namely Southeast, Central East, and South.All patients were residing in these cities at the time of their lung cancer (LC) diagnosis.
In seven of these cities (four in São Paulo, one in the Federal District, one in Paraná, and one in the State of Minas Gerais), accounting for a total of 18 cases, we searched for the official government environmental pollution data in the form of PM2.5 measurements from ground-based monitoring stations (Table 3) and available at the Air Quality Historical Data Platform website.Available online: https://aqicn.org/dataplatform/register/(accessed on 16 January 2023).These measurements ranged from 63.9 (in the city of São Paulo) to 30 (in the city of Brasília) and surpassed the PM2.5 levels specified as indicative of poor air quality by the European Ambient Air Quality Directives (as depicted by the horizontal red dotted line in Figure 6b) [17].
We visually depicted the distribution of LC cases on a map of Brazil, accentuating regions with elevated PM2.5 concentrations based on satellite imagery sourced from the IQAIR website [Available online: https://www.iqair.com(accessed on 16 January 2023)] (Figure 6a), which coincided with the locations of the patients' cities.

PM 2.5 Measurements
All cases were categorized across 17 cities in six Brazilian States (as outlined in Table 3), with the majority (n = 8) situated in the state of São Paulo, followed by Minas Gerais (n = 3), Paraná (n = 3), Federal District (n = 1), Goiás (n = 1), and Santa Catarina (n = 1).These cities correspond to three distinct regions within Brazil, namely Southeast, Central East, and South.All patients were residing in these cities at the time of their lung cancer (LC) diagnosis.
In seven of these cities (four in São Paulo, one in the Federal District, one in Paraná, and one in the State of Minas Gerais), accounting for a total of 18 cases, we searched for the official government environmental pollution data in the form of PM 2.5 measurements from ground-based monitoring stations (Table 3) and available at the Air Quality Historical Data Platform website.Available online: https://aqicn.org/data-platform/register/(accessed on 16 January 2023).These measurements ranged from 63.9 (in the city of São Paulo) to 30 (in the city of Brasília) and surpassed the PM 2.5 levels specified as indicative of poor air quality by the European Ambient Air Quality Directives (as depicted by the horizontal red dotted line in Figure 6b) [17].
We visually depicted the distribution of LC cases on a map of Brazil, accentuating regions with elevated PM 2.5 concentrations based on satellite imagery sourced from the IQAIR website [Available online: https://www.iqair.com(accessed on 16 January 2023)] (Figure 6a), which coincided with the locations of the patients' cities.From left to right, the vertical columns display the associated cities and states, along with the patients' ages and years of diagnosis.Subsequently, the mean daily PM2.5 values, methodologies, and the duration of measurement are presented.

Discussion
As previously described, the TP53 R337H Brazilian germline founder mutation had a general population incidence of 0.3% in southeastern and southern regions, with less penetrance besides a differential spectrum of tumors [13,[18][19][20].In a retrospective pan-

Discussion
As previously described, the TP53 R337H Brazilian germline founder mutation had a general population incidence of 0.3% in southeastern and southern regions, with less penetrance besides a differential spectrum of tumors [13,[18][19][20].In a retrospective pancancer analysis in Brazil, Sandoval and collaborators described a 2% incidence of this variant, highlighting the high prevalence in that territory [21].Characteristically, this variant is predominantly described as germline, with few cases reported as somatic [22,23], and some specialists recommend germinal evaluation when this alteration is found in somatic assays.Some studies conducted in Brazil have described an enrichment of lung adenocarcinomas (LUADs) in R337H carriers.Starting from R337H carriers, Barbosa et al. found an incidence of LUADs of 5.4%, with most of these having activating mutations in EGFR (eight of nine patients) [15].Another retrospective analysis of 513 patients with LC (83.8% LUADs) described TP53 mutation in 53.6% of the cases using somatic NGS assay, 2.3% of these corresponding to the R337H variant (12 patients), which had a statistical association to EGFR and HER2 alterations [16].However, a limited number of patients or clinical information did not allow for further conclusions in these studies.
Here, we describe the largest cohort of patients with LC in a TP53 R337H setting (28 patients), representing 16% of the 175 cases with diagnosed tumors from the BLISS database.In line with other series [15,16], we found a predominance of women (68%), with the age of above 50 years old at diagnosis (mean of 54.6); all cases were adenocarcinoma, in nonsmokers (78.6%), and with alterations in the ERBB family (70% of the tested cases; 14 of 20 patients) concentrated in exons 18 to 20 (to the EGFR), and one case with HER 2 exon 20 insertion.Interestingly, LUAD was the first tumor diagnosed in 78.6% of cases, which was not previously described.These findings reinforce the possibility of interactions between the TP53 R337H variant and the ERBB family in LUAD carcinogenesis; however, more studies are needed to elucidate the possible mechanism of these findings.Most diagnoses were in the context of an initially resectable disease, which may be due to the profile of patients seen in a private service, where the search for medical assistance occurs earlier, or due to incidental findings in imaging tests.Because of the small number of cases, we did not perform statistical analysis, and the clinical outcomes were only described.For exposure, none of the patients with localized disease who underwent surgery died (n = 15), and they had a median disease-free survival of 106.9 months.In patients with metastatic disease and activating EGFR mutations treated with EGFR inhibitors, there was no numerical difference in progression-free survival between patients on first-generation (erlotinib and gefitinib) or third-generation (osimertinib) EGFR inhibitors.
A previous analysis of a large patient population found a positive association between air pollutants and tumor incidence, such as brain tumors [23] and lung cancer [24], with emphasis on chronic high levels of PM 2.5 in µg/m 3 [6].Experimentally, the exposition of human bronchial epithelial cell lines to airborne particulate matter (PM) results in an inflammatory response as dose-dependent reduced cell viability and increased reactive oxygen species production [25].An elegant study published by Hill et al. proved that exposing airway cells that already harbor driver mutations, such as EGFR and KRAS, to PM 2.5 would induce a stem cell phenotype that, ultimately, results in the formation of adenocarcinomas.In this carcinogenesis model, the production of IL-1ß by infiltrating macrophages would be a pivotal step in the process since the use of this cytokine antagonist was able to inhibit tumor formation [26].
According to this, we found high PM 2.5 levels in the cities of the patients in our study (Table 3 and Figure 6b), with mean daily levels above the threshold designated as poor by the European Air Quality Index, and the LUAD diagnosis in TP53 R337H context concentrate in areas of more pollution levels in Brazilian map (Figure 6a).In addition, in all cases (n = 16) with somatic NGS assessment, a driver mutation was found (besides the TP53 R337H), which could reinforce the hypothesis that pollution exposure could promote cancer development in already initiated bronchial cells [27].Finally, some TP53 variants, but not the normal gene, could enhance the IL-1ß activity by suppressing the production of secreted interleukin-1 receptor antagonists in cancer cell and xenograft models [28].Future studies are needed to elucidate the possible mechanisms of interaction between the TP53 R337H variant and IL-1β signaling in the context of exposure to high levels of PM 2.5 in the carcinogenesis of lung adenocarcinoma.
The lack of somatic NGS results is a significant limitation of this study.Furthermore, the absence of data regarding the duration of residence in cities with elevated levels of PM 2.5 prior to the tumor diagnosis, and the lack of a control group carrying the R337H variant without the occurrence of lung cancer development, render the specific correlation between air pollution and LUAD carcinogenesis a matter of mere speculation.Further studies are needed to address this inquiry.However, we assembled the largest cohort of patients with LUADs in the LFS R337H context, describing the clinical characteristics, clinical outcomes, and tumor somatic molecular profile in more detail, in addition to formulating hypotheses for the role of TP53 R337H in the carcinogenesis of these tumors.

Demographic, Clinical, and Genomic Information
We conducted a single-center retrospective analysis of 370 individuals enrolled in the Brazilian Li-Fraumeni Syndrome Study (BLISS) registry at Sírio-Libanes Hospital, São Paulo City, Brazil [16].All individuals harbored a TP53 R337H germline pathogenic variant, which was confirmed through commercial germline testing methods, including Sanger sequencing, multiplex ligation-dependent probe amplification (MLPA), or nextgeneration sequencing platforms, all of which are covered by private health insurance in our country.Demographic and clinical data, including age at diagnosis, gender, smoking history, initial tumor diagnosis, history of other neoplasms, lung cancer histology, and staging at diagnosis, were collected for patients treated between 2010 and 2022, and they are presented in tables and diagrams based on variable characteristics.Furthermore, we documented the presence and methodology of somatic molecular tumor assessments in a table and diagram, highlighting the molecular profiles.

Clinical Outcomes
Using the Kaplan-Meier method, we established the medians of disease-free survival (time from definitive treatment to the recurrence of tumor or death due to any cause), progression-free survival (time from the beginning of systemic therapy to documented progression or death due to any cause), and overall survival (time from diagnosis to death due to any reason).Data were censored after the last present evaluation in the reference center for patients who did not have events.The Kaplan-Meier curves were generated by using GraphPad Prism version 8.

Measurement of Pollution Level
To investigate the potential correlation between environmental pollution and the incidence of lung cancer, we determined the daily mean PM 2.5 levels, measured in µg/m 3 , using the data of the respective cities of the patients.These measurements were obtained using ground-level monitoring devices, and the data sources were the official government environmental agencies of each state, which can be accessed at the Air Quality Historical Data Platform website.Available online: https://aqicn.org/data-platform/register/(accessed on 16 January 2023).The PM 2.5 values were then depicted using box plots and compared against the air quality standards set forth by the European Union [18].Logarithmic transformations were applied to these data to enhance graphical representation.

Conclusions
Individuals carrying the Brazilian founder mutation TP53 R337H exhibit a heightened incidence of lung cancer compared with more prevalent global variants.Within the TP53 R337H context, lung cancer may manifest as the primary indication of Li-Fraumeni syndrome, predominantly diagnosed in never-smoking women over the age of 50.All the observed cases were adenocarcinomas, with the majority featuring driver mutations highlighted by actionable alterations in the ERBB family.Descriptively, it appears that these tumors may yield benefits from systemic therapies in advanced settings.As a speculative hypothesis, our data indicate a potential correlation between elevated environmental pollu-

11 Figure 1 .
Figure 1.Patients included in the BLISS database: (A) proportion of patients with cancer diagnosis; (B) proportion of lung cancer from all tumors.

Figure 1 .
Figure 1.Patients included in the BLISS database: (A) proportion of patients with cancer diagnosis; (B) proportion of lung cancer from all tumors.

Figure 2 .
Figure 2. Genomic landscape of lung adenocarcinomas.Genomic profile of the 20 patients submitted to tumoral multigenic next-generation sequencing (NGS) platforms or real-time PCR (rtPCR) for EGFR and ALK.NGS was performed in 16 patients and rtPCR was performed in 4 (indicated by *).Most cases indicated mutations in driver genes.A total of 23 alterations, comprising 7 actionable and 4 related to systemic treatment resistance, were described.In horizontal lines, individual characteristics of the patients are listed, such as average levels of exposure to PM2.5, gender, smoking, and evaluated genes.Each column indicates an individual patient.On the right, the genetic alterations are described and illustrated with colored squares.Abbreviations: del-deletion; indel-insertion or deletion; inser-insertion; PM2.5-particulate matter < 2.5mm of diameter.

Figure 2 .Figure 3 .
Figure 2. Genomic landscape of lung adenocarcinomas.Genomic profile of the 20 patients submitted to tumoral multigenic next-generation sequencing (NGS) platforms or real-time PCR (rtPCR) for EGFR and ALK.NGS was performed in 16 patients and rtPCR was performed in 4 (indicated by *).Most cases indicated mutations in driver genes.A total of 23 alterations, comprising 7 actionable and 4 related to systemic treatment resistance, were described.In horizontal lines, individual characteristics of the patients are listed, such as average levels of exposure to PM 2.5 , gender, smoking, and evaluated genes.Each column indicates an individual patient.On the right, the genetic alterations are described and illustrated with colored squares.Abbreviations: del-deletion; indel-insertion or deletion; inser-insertion; PM 2.5 -particulate matter < 2.5mm of diameter.Int.J. Mol.Sci.2023, 24, x FOR PEER REVIEW 5

Figure 3 .
Figure 3. Clinical outcomes of 16 patients with early disease submitted to surgery: (a) median disease-free survival of the sixteen patients assessed; (b) overall survival of these patients.None of them had died due to LC. Abbreviation: NR-not reached.For individuals with advanced disease and EGFR mutations (n = 11), the median progression-free survival was 19.6 and 17.4 months for the first-generation (erlotinib and gefitinib) and third-generation (osimertinib) EGFR inhibitors, respectively, as illustrated in Figure 4a,b.In this context, the median overall survival was not reached, as shown in Figure 4c,d.Notably, one death occurred during osimertinib treatment, coinciding with the emergence of BRAF amplification.

Figure 3 .
Figure 3. Clinical outcomes of 16 patients with early disease submitted to surgery: (a) disease-free survival of the sixteen patients assessed; (b) overall survival of these patients.N them had died due to LC. Abbreviation: NR-not reached.

Figure 4 .
Figure 4. Clinical outcomes in 11 patients with advanced disease treated with EGFR inhibi median progression-free survival on first-generation EGFR inhibitor treatment (n = 6; follow-up of 15.7 months); (b) median progression-free survival on third-generation EGFR in treatment (n = 5; median of follow-up of 12 months); (c) median overall survival on first-gen EGFR inhibitor treatment (n = 6; median of follow-up of 50.3 months); (d) median overall s on third-generation EGFR inhibitor treatment (n = 5; median of follow-up of 67.6 months).

Figure 4 .
Figure 4. Clinical outcomes in 11 patients with advanced disease treated with EGFR inhibitors: (a) median progression-free survival on first-generation EGFR inhibitor treatment (n = 6; median follow-up of 15.7 months); (b) median progression-free survival on third-generation EGFR inhibitor treatment (n = 5; median of follow-up of 12 months); (c) median overall survival on first-generation EGFR inhibitor treatment (n = 6; median of follow-up of 50.3 months); (d) median overall survival on third-generation EGFR inhibitor treatment (n = 5; median of follow-up of 67.6 months).

Figure 5 .
Figure 5. Clinical evolution on systemic treatments of patients #6 and #11: (a) a 41-year-old man with metastatic LC to bone and lungs with EGFR exon 20 insertion was submitted to six lines of treatment with clinical and radiologic benefit until multisystemic progression after seven months of an experimental EGFR oral inhibitor for exon 20 alterations; (b) a 33-year-old man with previous resected LC harboring, HER 2 exon 20 insertion, and submitted to four cycles of platinum doublet experienced a hepatic recurrence 13 months after.Then, he received three lines of systemic treatment, as represented above, with clinical and radiographic benefits for 25 months until CNS progression.Abbreviations: carbo-carboplatin; CDDP-cisplatin; pmtx-pemetrexed; TDM-1trastuzumab-emtansine; x followed by a number indicates the number of treatment cycles.

Figure 5 .
Figure 5. Clinical evolution on systemic treatments of patients #6 and #11: (a) a 41-year-old man with metastatic LC to bone and lungs with EGFR exon 20 insertion was submitted to six lines of treatment with clinical and radiologic benefit until multisystemic progression after seven months of an experimental EGFR oral inhibitor for exon 20 alterations; (b) a 33-year-old man with previous resected LC harboring, HER 2 exon 20 insertion, and submitted to four cycles of platinum doublet experienced a hepatic recurrence 13 months after.Then, he received three lines of systemic treatment, as represented above, with clinical and radiographic benefits for 25 months until CNS progression.Abbreviations: carbo-carboplatin; CDDP-cisplatin; pmtx-pemetrexed; TDM-1-trastuzumabemtansine; x followed by a number indicates the number of treatment cycles.
right, the vertical columns display the associated cities and states, along with the patients' ages and years of diagnosis.Subsequently, the mean daily PM 2.5 values, methodologies, and the duration of measurement are presented.

Figure 6 .
Figure 6.Distribution of cases in Brazil and average daily values of PM2.5:(a) location of patients diagnosed with LC according to the PM2.5 heat map obtained via satellite (the redder the colors, the higher the PM2.5 value); (b) daily mean values are represented in box plots of cities with cases and comparison with the value established by the European Society for the Environment as poor air quality (EAQI poor).We applied logarithmic scaling to the values for better graphical representation.Note that most values are above the levels designated as EAQI poor.

Figure 6 .
Figure 6.Distribution of cases in Brazil and average daily values of PM 2.5 : (a) location of patients diagnosed with LC according to the PM 2.5 heat map obtained via satellite (the redder the colors, the higher the PM 2.5 value); (b) daily mean values are represented in box plots of cities with cases and comparison with the value established by the European Society for the Environment as poor air quality (EAQI poor).We applied logarithmic scaling to the values for better graphical representation.Note that most values are above the levels designated as EAQI poor.

Table 1 .
Demographic and clinical of LC LFS R337H patients.

Table 2 .
ERRB family's mutational profile of LC LFS R337H patients who performed some tumoral molecular assessment.

Table 3 .
Cities and states distribution of LC cases with levels, methods, and periods for measurement of PM 2.5 levels.

Table 3 .
Cities and states distribution of LC cases with levels, methods, and periods for measurement of PM2.5 levels.