Abstract
Objective:
The aim of this study is to describe four of the most common autoantibodies against components of the Th/To complex: hPOP1, RPP25, RPP30, and RPP40. We report their prevalence and clinical characteristics in a systemic sclerosis (SSc) population, and determine whether these specificities associate with cancer.
Methods:
A case control study was performed utilizing data from the Johns Hopkins Scleroderma Center Cohort. A total of 804 adult patients with SSc were included; 401 SSc patients with no history of cancer after at least 5 years of disease were compared to 403 SSc patients who ever had a history of cancer. Antibodies against hPOP1, RPP25, RPP30, and RPP40 were assayed by immunoprecipitation of 35S-methionine-labeled proteins generated by in vitro transcription/translation. Demographic and clinical characteristics were compared between groups.
Results:
67/804 (8.3%) of patients had antibodies against any component of the Th/To complex. Patients with antibodies to any component were significantly more likely to have limited cutaneous disease, less likely to have tendon friction rubs, and more likely to have findings consistent with interstitial lung disease or pulmonary hypertension. Patients with antibodies against hPOP1, RPP25, RPP30, and/or RPP40 were significantly less likely to develop cancer within 2 years of SSc-onset (0% vs 11%, p=0.016).
Conclusion:
SSc patients who produce autoantibodies to components of the Th/To complex have a clinical phenotype characterized by limited cutaneous disease and pulmonary involvement. Our findings show that the presence of any Th/To autoantibody may have a protective effect against contemporaneous cancer.
Introduction:
Important new insights into the relationship between systemic sclerosis (scleroderma, SSc) and cancer have emerged over the past decade. Recent research has highlighted that both time and autoantibodies are critical filters in risk-stratifying SSc patients for malignancy [1–2]. Whereas some autoantibody specificities have been demonstrated to confer an increased risk of cancer around the time of SSc symptom onset (eg, anti-RNApol3 or anti-RNPC3), others have suggested a protective role (eg, anti-centromere or the combined presence of anti-RNApol3+anti-RPA194) [2–4]. Furthermore, our group has observed that SSc patients lacking antibodies against centromere, topoisomerase 1, and RNApol3 may have an increased risk of cancer-associated scleroderma [1,4]. To address the important question of whether there are additional SSc antibody specificities that associate with cancer (either protectively or as markers of increased risk), we developed a large, meticulously phenotyped case control cohort of SSc patients with and without malignancy. In this report, we investigated the relationship of antibodies against four well-defined components of the Th/To complex and SSc/cancer status.
The ribonuclease mitochondrial RNA processing complex (RNase MRP) and ribonuclease P (RNase P) are both enzymes that are involved in processing ribosomal, mitochondrial and transfer RNA. In the early 1980s, sera from SSc patients were found to have autoantibodies directed against these complexes [5–6], and the nomenclature evolved to designate “anti-Th” for RNase MRP and “anti-To” for RNase P. In subsequent decades, at least 9 individual proteins associated with each of these complexes have been isolated and characterized (hPOP1, hPOP4, hPOP5, RPP14, RPP20, RPP21, RPP25, RPP30, and RPP40) [7–8]. Many components have been described as targets of autoantibodies in SSc, and are collectively referred to as anti-Th/To antibodies. Prior studies have concluded the prevalence of these antibodies in SSc patients is generally low (5–15%), and the clinical phenotype is consistent with the CREST syndrome or limited cutaneous SSc [9–11]. However, these studies were performed in relatively small SSc cohorts and have not assessed autoantibodies to individual components of the Th/To complex. Furthermore, how the components relate to one another and their association with cancer has not been explored to date.
In the current manuscript, we describe the prevalence and clinical characteristics of SSc patients tested for four of the most common autoantibodies of the Th/To complex: hPOP1, RPP25, RPP30, and RPP40. We find that patients with autoantibodies to any of these four components of the Th/To complex are protected from cancer-associated SSc and possess a clinical phenotype most notable for pulmonary involvement.
Methods
Patients –
A case control study was performed utilizing data and banked sera from the Johns Hopkins Scleroderma Center Cohort. A total of 804 adult patients with SSc were included, 97% of whom met 2013 Systemic Sclerosis ACR/EULAR Classification Criteria. A total of 401 SSc patients who had no history of cancer after at least 5 years of disease (a median follow-up of 14 years (IQR 9–24)), were compared to 403 SSc patients who ever had a history of cancer. Patient reported cancer diagnoses were validated by review of medical records and pathology reports when available. Of the 403 patients with cancer, 94 had a non-melanotic skin cancer (basal and squamous cell carcinomas). The time interval between cancer diagnosis and SSc onset, defined by the first of either Raynaud’s onset or the first non-Raynaud’s symptom, was calculated. The cancer-SSc interval was further binned in 3 time windows (± 2, 3, or 5 years) that may suggest the presence of cancer-associated SSc, as previously defined [4, 12]. The following clinical variables are prospectively collected in the Johns Hopkins Scleroderma Center Database: Medsger organ-specific scleroderma severity scores [13], modified Rodnan skin score, echocardiogram data including ejection fraction (EF) and right ventricular systolic pressure (RVSP), pulmonary function testing data including forced vital capacity (FVC) and diffusing capacity (DLCO) (standardized for age and sex and reported as the percentage of the predicted value [14–15]), cutaneous subtype [16], presence/history of scleroderma renal crisis, tendon friction rubs, myopathy, and synovitis. Patients were referred for right heart catheterization (RHC) at the discretion of the treating rheumatologist as previously described [17]. Pulmonary hypertension (PH) was defined as a mean pulmonary artery pressure (mPAP) ≥ 21 mm Hg obtained by RHC as recently recommended by the 6th World Symposium on Pulmonary Hypertension [18]. Sensitivity analyses were performed using the prior cut-off of mPAP ≥ 25 mmHg. The Johns Hopkins IRB approved this study protocol.
Autoantibody assays –
Complementary DNAs encoding full-length FLAG-tagged human POP1, RPP40, RPP30 and RPP25 were cloned (RPP30) or purchased (RPP40 and POP1 from Origene, Rockville MD). DNAs were sequence verified, and used to generate 35S-methionine–labeled proteins by in vitro transcription/translation (IVTT) reactions, per the manufacturer’s protocol (Promega). Immunoprecipitations (IPs) were performed using these products as input; the washed IPs were electrophoresed on 10% SDS-PAGE gels and visualized by fluorography [19]. An anti-FLAG IP was included in each sample set and fluorogram exposures were standardized to give anti-FLAG IP bands of similar intensity. All IPs were quantitated by densitometric scanning and normalized to the FLAG IP reference included in the set. Sera from 34 healthy controls banked at the Johns Hopkins site were also tested by IP with each of the IVTT products; an anti-FLAG IP was included in each of these control sets as a reference calibrator. No IP band was detected with any of the healthy control sera (see Fig 1 for representative examples). The cutoff for antibody positivity was assigned a calibrated OD value of 5 units, which corresponded to the faintest reproducible IVTT-IP band detectable by eye (calibrated OD values (“units”) ranged from 5–127). Representative examples of IPs using each of the IVTT products are shown in Fig 1. Cutoffs were defined as: <5 units, neg; 5–15 units, +; >15–50 units, ++ and >50 units, +++.
Figure 1: Immunoprecipitation (IP) assay to identify antibodies against 4 different components of the Th/To complex.
Sera from 6 scleroderma patients (lanes 2–7) and 3 healthy controls (lanes 8–10) were used to IP 35S-methionine-labeled hPOP1, RPP40, RPP30 or RPP25 generated by in vitro transcription/translation. hPOP1 antibodies were detected in all 7 of the scleroderma patients shown in the figure; RPP40 antibodies were found in 6 (absent in #7), RPP30 antibodies were present in 4 of these sera (absent in #5 and #7) and RPP 25 antibodies were in 4 of the sera (absent in # 6 and #7). An anti-FLAG calibrator IP, performed using an anti-FLAG monoclonal antibody (Sigma), is shown in lane 1 of each set.
Given prior studies demonstrating that the risk of cancer-associated SSc may be modified by the presence of additional immune responses, we were interested in testing whether positivity to any of the anti-Th/To components conferred cancer protection among patients with anti-RNApol3 antibodies. Anti-RNApol3 antibody status was determined by line blot assays using the Euroimmun platform (Euroimmun Diagnostics, Lubeck, Germany).
Statistical Analyses –
Statistical analysis was performed using Stata software Version 14 (College Station, TX). The primary analysis excluded non-melanotic skin cancers (basal and squamous cell carcinomas), with secondary analyses including all cancers. For comparison between groups, Fisher’s exact was utilized for categorical/binary variables and rank sum test for comparison between continuous variables given a non-normal distribution. Medsger organ-specific severity scores were analyzed as an ordinal variable using the Kruskal-Wallis test. A p-value <0.05 was considered statistically significant, adjusting for multiple comparison testing.
Results
Demographics and clinical characteristics of scleroderma patients with or without cancer
The majority of patients were Caucasian females with SSc-disease duration of 5.9±11 years (Table 1). Comparing SSc patients with a history of cancer (excluding non-melanotic skin cancer) to those without, patients with cancer were significantly less likely to be Hispanic (2% vs 9%, p=0.002), more likely to be Caucasian (89% vs 73%, p<0.0001), and older at SSc-symptom onset (median 48.2 [IQR 38–58] vs 39.5 [IQR 28–49], p<0.0001). SSc patients with cancer had an overall higher maximum MRSS (median 9 [IQR 4–20] vs 6 [IQR 3–17], p=0.003) and had a higher mortality (44% vs 31%, p=0.001). Medsger organ specific severity scores were comparable between groups, with the notable exception of Raynaud’s severity score, which was less severe in the cancer group (p=0.002). Additional demographic and clinical variables are depicted in Table 1. After adjustment for multiple comparisons (p=0.05/30 variables equates to a significance threshold of <0.002), only ethnicity, race, and death remained statistically significant. Similar patterns were observed when comparing SSc patients with any cancer (including non-melanotic skin cancers, n=403) to patients without cancer with regards to race, ethnicity, and age of SSc onset (Supplemental Table 1). With the inclusion of non-melanotic skin cancers, mortality was no longer significantly different, and patients with cancer had a significantly shorter disease duration (5 vs 6.5 years, p<0.004).
Table 1.
Demographics and SSc Characteristics comparing patients with a history of cancer to those who have never had a cancer. Patients with cancer history are more likely to be Caucasian, older at SSc onset, and have a higher Modified Rodnan Skin Score (MRSS).
| Patient Characteristics | No History of Cancer (N=401) | *Cancer History (N=309) | p-value |
|---|---|---|---|
| n (%) / median (IQR) | n (%) / median (IQR) | ||
| Sex | 345 (86) | 254 (82) | 0.1760 |
| Hispanic Ethnicity | 18 (9) | 3 (2) | 0.0020 |
| Race | |||
| Caucasian | 292 (73) | 277 (89) | <0.0001 |
| African American | 81 (20) | 23 (7) | |
| History of smoking (ever) | 185 (47) | 161 (52) | 0.1490 |
| Age at SSc-symptom onset (RP or non-RP) (IQR) | 39.5 (28–49) | 48.2 (38–58) | <0.0001 |
| Meets ACR 2013 Criteria | 387 (97) | 300 (97) | 1.0000 |
| Follow-up Duration (yrs) | 14.4 (9–24) | 12.6 (6–23) | 0.0969 |
| Disease Duration Upon Cohort Entry | 6.5 (2.6–15) | 4.9 (1.5–15) | 0.0740 |
| Cutaneous Subtype | |||
| Sine | 13 (3) | 9 (3) | 0.2730 |
| Limited | 214 (53) | 144 (46) | |
| Type II Skin | 36 (9) | 33 (11) | |
| Diffuse | 138 (34) | 124 (40) | |
| Cutaneous Subtype (limited vs diffuse) | 138 (34) | 124 (40) | 0.1170 |
| Maximum Modified Rodnan Skin Score | 6 (3–17) | 9 (4–20) | 0.0030 |
| History of scleroderma renal crisis | 16 (4) | 19 (6) | 0.2220 |
| Minimum FVC % predicted over follow-up (n=668/710) | 70 (54–84) | 70 (57–84) | 1.0000 |
| Maximum Right Ventricular Systolic Pressure (n=590/710) | 37 (32–50) | 40 (33–53) | 0.0700 |
| Pulmonary Hypertension (mean PAP>20) (n=148) | 60 (15) | 41 (13) | 0.5880 |
| Pulmonary Hypertension (mean PAP>24) (n=148) | 54 (13) | 37 (12) | 0.5730 |
| Presence of baseline TFR | 35 (9) | 41 (13) | 0.0660 |
| Presence of synovitis on any exam | 84 (21) | 68 (22) | 0.7820 |
| Deceased | 124 (31) | 131 (44) | 0.0010 |
| Serology | |||
| Any ThTo Component (hPOP1, RPP25 RPP30, or RPP40) | 36 (9) | 23 (7) | 0.4950 |
| hPOP1 | 36 (9) | 22 (7) | 0.4090 |
| RPP25 | 23 (6) | 12 (4) | 0.2970 |
| RPP30 | 24 (6) | 20 (6) | 0.8760 |
| RPP40 | 31 (8) | 23 (7) | 1.0000 |
RP=Raynaud’s phenomenon, SSc=systemic sclerosis, TFR=tendon friction rubs.
Excluding non-melanotic skin cancers
A total of 67/804 (8.3%) of patients had antibodies against any component of the Th/To complex, with no differences in frequency noted between the cancer and no cancer groups (Table 1). Of the 67 anti-Th/To-positive patients, 10/67 (15%) had antibodies against 1 or 2 components, while the majority (57/67, 85%) targeted 3 or 4 components (Table 2). Given that so few patients produced autoantibodies to only 1 or 2 components, we explored whether these were borderline/low positives. Of the 6 patients with antibodies against only 1 Th/To subunit (5 anti-hPOP1 and 1 anti-RPP40), 4 were moderate/high titer (all were anti-hPOP1).
Table 2.
Number of Th/To subunit specificities within individual patients. The majority of patients have antibodies against multiple (>2) subunits. Of the 6 patients with antibodies against only 1 Th/To subunit (5 with anti-hPOP1 and 1 with anti-RPP40), most are high-titer positives (++/+++): 4/5 anti-hPOP1 ++/+++ and 0/1 anti-RPP40.
| Number of RPP Subunits (hPOP1, RPP25, RPP30, RPP40) | # of Pts | % Cohort |
|---|---|---|
| 0 | 737 | 91.67 |
| 1 | 6 | 0.75 |
| 2 | 4 | 0.50 |
| 3 | 25 | 3.11 |
| 4 | 32 | 3.98 |
| Total | 804 | 100.00 |
Clinical phenotype of patients with antibodies against different Th/To components
Patients who were positive for each Th/To antibody specificity (hPOP1, RPP25, RPP30, and RPP40) were compared to all other SSc patients in the cohort with regards to demographics and SSc characteristics (Table 3). Patients with antibodies to any of the four components were more likely to have a smoking history, limited cutaneous disease and thus a lower median maximum MRSS, and a lower prevalence of tendon friction rubs.
Table 3.
Demographic and clinical features by prevalence of antibodies against various subunits of the Th/To complex. Patients with antibodies against any subunit tend to have limited cutaneous disease, a lower MRSS, and fewer TFR.
| hPOP1 N (%)/median (IQR) Negative (N=739) | Positive (N=65) | p-value | RPP25 N (%)/median (IQR) Negative (N=762) | Positive (N=42) | p-value | RPP30 N (%)/median (IQR) Negative (N=756) | Positive (N=48) | p-value | RPP40 N (%)/median (IQR) Negative (N=742) | Positive (N=62) | p-value | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Female Sex | 54 (83) | 621 (84) | 0.8600 | 640 (84) | 35 (83) | 0.8310 | 635 (84) | 40 (83) | 0.8410 | 624 (84) | 51 (82) | 0.7190 |
| Ethnicity | 1 (3) | 21 (5) | 1.0000 | 21 (5) | 1 (5) | 0.6130 | 21 (5) | 1 (5) | 1.0000 | 21 (5) | 1 (3) | 1.0000 |
| Caucasian | 52 (80) | 609 (82) | 0.8860 | 628 (82) | 33 (79) | 0.7460 | 621 (82) | 40(83) | 0.8880 | 609 (82) | 52 (84) | 0.9440 |
| African American | 11(17) | 94 (13) | 97 (13) | 8 (19) | 98 (13) | 7 (15) | 96 (13) | 9 (15) | ||||
| History of smoking (ever) | 347(48) | 39 (61) | 0.0500 | 358 (48) | 28 (68) | 0.0100 | 356 (48) | 30 (64) | 0.0350 | 348 (47) | 38 (62) | 0.0320 |
| Age at SSc-symptom onset (RP or non-RP) | 44 (33–55) | 42 (32–48) | 0.1200 | 44 (33–55) | 43 (37–51) | 0.6360 | 44 (33–55) | 44 (32–54) | 0.8740 | 44 (33–55) | 43 (32–53) | 0.5120 |
| Disease Duration Upon Cohort Entry | 5.6 (1.9–15) | 9.8 (4.2–16) | 0.1150 | 5.7 (1.9–15) | 9.8 (4.4–15) | 0.0760 | 6 (1.9–15) | 10 (3.6–16) | 0.0960 | 5.5 (1.9–15) | 10 (4.4–16) | 0.0230 |
| Cutaneous Subtype | ||||||||||||
| Sine | 26 (4) | 1 (2) | <0.0001 | 26 (3) | 1 (2) | <0.0001 | 26 (3) | 1 (2) | <0.0001 | 26 (4) | 1 (2) | <0.0001 |
| Limited | 356 (48) | 53 (82) | 376 (49) | 33 (79) | 369 (49) | 40 (83) | 358 (48) | 51 (82) | ||||
| Type II Skin | 74 (10) | 5 (8) | 74 (10) | 5 (12) | 74 (10) | 5 (10) | 74 (10) | 5 (8) | ||||
| Diffuse | 283 (38) | 6 (9) | 286 (38) | 3 (7) | 287 (38) | 2 (4) | 284 (38) | 5 (8) | ||||
| Cutaneous Subtype (Limited vs Diffuse) | 283 (38) | 6 (9) | <0.0001 | 286 (38) | 3 (7) | <0.0001 | 287 (38) | 2 (4) | <0.0001 | 284 (38) | 5 (8) | <0.0001 |
| Maximum MRSS | 8(4–19) | 4 (3–6) | <0.0001 | 7 (4–19) | 4 (2–6) | <0.0001 | 7 (4–19) | 4 (3–6) | <0.0001 | 8 (4–19) | 4 (3–6) | <0.0001 |
| History of scleroderma renal crisis | 39 (5) | 1 (2) | 0.2430 | 39 (5) | 1 (2) | 0.7160 | 39 (5) | 1 (2) | 0.5050 | 39 (5) | 1 (2) | 0.3560 |
| Minimum FVC % predicted over follow-up | 72 (56–86) | 69 (47–81) | 0.7910 | 72 (56–85) | 70 (52–81) | 0.7450 | 72 (56–86) | 66 (42–81) | 0.4420 | 72 (56–86) | 70 (46–82) | 0.6840 |
| Maximum Right Ventricular Systolic Pressure (mmHg) | 38 (32–50) | 40 (34–63) | 0.2020 | 38 (32–50) | 41 (34–65) | 0.3500 | 38 (32–50) | 42 (34–66) | 0.2250 | 38 (32–50) | 40 (34–63) | 0.1140 |
| Pulmonary Hypertension (mean PAP>20) | 100 (14) | 15 (23) | 0.0420 | 107 (14) | 8 (19) | 0.3650 | 104 (14) | 11 (23) | 0.0890 | 101 (14) | 14 (23) | 0.0590 |
| Pulmonary Hypertension (mean PAP>24) | 87 (12) | 14 (22) | 0.0310 | 93 (12) | 8 (20) | 0.2270 | 90 (12) | 11 (23) | 0.0400 | 88 (12) | 13 (21) | 0.0460 |
| Baseline TFR | 81 (11) | 0 (0) | 0.0010 | 81 (11) | 0 (0) | 0.0160 | 81 (11) | 0 (0) | 0.0110 | 81 (11) | 0 (0) | 0.0020 |
| Presence of synovitis on any exam | 153 (21) | 14 (22) | 0.8740 | 158 (21) | 9 (21) | 0.8480 | 156 (21) | 11 (23) | 0.7140 | 155 (21) | 12 (19) | 0.8710 |
| Deceased | 246 (34) | 24 (38) | 0.5830 | 256 (34) | 14 (34) | 1.0000 | 248 (33) | 22 (46) | 0.0850 | 246 (34) | 24 (39) | 0.4000 |
MRSS=Modified Rodnan Skin Score, RVSP=right ventricular systolic pressure, FVC = forced vital capacity, TFR= tendon friction rub, P-value: Fisher’s exact for binary and categorical variable comparisons; rank sum for non-parametric continuous variable comparisons.
Upon analyzing Medsger Severity Scores for individual organ systems and anti-hPOP1 antibodies, significant associations were observed for less severe GI disease (p=0.012) and more severe lung disease (p=0.011). Similarly, patients with antibodies against RPP40 had less severe GI disease (p=0.034) and more severe lung disease (p=0.016). Anti-RPP30-positive patients were more likely to have severe lung disease (p=0.020). No associations were observed for RPP25 antibodies.
Given the absence of comprehensive high-resolution CT scan (HRCT) data in our cohort, we focused on Medsger Lung Severity Scores (which incorporate PFT parameters and need for supplemental oxygen) as well as right heart catheterization data. Patients with antibodies to any of the Th/To components were more likely to have worse pulmonary disease as defined by a higher maximum Medsger Lung Severity Score (Kruskal-Wallis p=0.013). When analyzing for the presence of severe lung disease (defined by a FVC or DLCO <50% of predicted or oxygen use required), patients with antibodies against hPOP1, RPP30, or RPP40 had significantly worse lung severity: 60% vs 37% for hPOP1 (p=0.001), 61% vs 37% for RPP30 (p=0.002), and 59% vs 37% for RPP40 (p=0.001). When examining FVC alone as a surrogate for ILD, patients with antibodies to any of the Th/To components were more likely to have severe restriction (FVC<50%) compared to those without antibodies to Th/To components: 27% vs 17%, p=0.088. Only anti-hPOP1-positive patients were more likely to develop pulmonary hypertension (23% v 14%, p=0.042). Upon raising the threshold for defining PH to a mean PAP≥25 mmHg, anti-hPOP1, -RPP30 or -RPP40-positive patients were more likely to develop pulmonary hypertension (22% vs 12% for hPOP1, p=0.03, 23% vs 12% for RPP30, p=0.04, and 21% vs 12% for RPP40, p=0.04).
Cancer risk and antibodies against Th/To components
Amongst all patients with autoantibodies to at least one component of the Th/To complex (n=67), 31 patients had a total of 38 cancers (including NMSC). Excluding NMSC, 23 patients had a cancer history (with 26 cancers). The presence of antibodies to Th/To components (hPOP1, RPP25, RPP30, and RPP40) conferred a protective effect for cancer-associated scleroderma (excluding NMSCs from analysis) (Table 4). Patients with antibodies against hPOP1, RPP25, RPP30, and/or RPP40 were significantly less likely to develop cancer within 2 years of SSc-onset (0% vs 11%, p=0.016). When examining the cancer-scleroderma interval amongst all patients with cancer, for antibodies against all of the components except RPP25, as the interval between SSc-onset and cancer shortened, the protective effect increased. Indeed, no patient with antibodies against any of the four components developed cancer within 2 years of SSc onset. Upon performing secondary analyses including all cancers (that is, including non-melanotic skin cancer), the magnitude of protection lessened, with only anti-RPP40 antibodies remaining statistically significant as a protective factor for cancer diagnosed within 2 years of SSc-symptom onset (3% vs 18%, p=0.04, Supplemental Table 2). We were unable to comment on a dose-response relationship (e.g. a decreasing prevalence of cancer with an increasing number of autoantibodies to Th/To components) because in the 10 patients with 1 or 2 autoantibodies to Th/To components, only one developed cancer (Supplemental Table 3).
Table 4.
Prevalence of cancer stratified by antibodies against individual Th/To subunits. In general, antibodies against Th/To subunit(s) appear to have a protective effect against cancer, with shorter-interval cancers being the most protected. Some analyses are limited by too few cancers in the anti-Th/To+ group, preventing conclusions about the relationship between the number of anti-Th/To specificities (epitope spreading) and cancer risk. This analysis excludes non-melanotic skin cancers; other analyses treating NMSC as cancers or NMSC in the cancer-negative group generate similar findings.
| Any Component | P0P1 | RPP25 | RPP30 | RPP40 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| POS n(%) | NEG n(%) | p-value | POS n(%) | NEG n(%) | p-value | POS n m | NEG n(%) | p-value | POS n(%) | NEG n m | p-value | POS n(%) | NEG n(%) | p-value | |
| Cancer (ever) | 23 (34) | 286(39) | 0.514 | 22(34) | 287(39) | 0.506 | 12(29) | 297(39) | 0.196 | 20(42) | 289 (38) | 0.648 | 23(37) | 286(39) | 0.892 |
| Cancer (within 5 years) | 4(8) | 110(20) | 0.055 | 4(8.5) | 110(20) | 0.078 | 1(3) | 113(20) | 0.018 | 4(13) | 110(19) | 0.486 | 4(9) | 110(19) | 0.109 |
| Cancer (within 3 years) | 2(2) | 83(16) | 0.048 | 2(4) | 83(16) | 0.047 | 1(3) | 84(15) | 0.069 | 2(7) | 83 (15) | 0.290 | 2(5) | 83(15) | 0.069 |
| Cancer (within 2 years) | 0(0) | 58(11) | 0.009 | 0(0) | 58(11) | 0.016 | 0(0) | 58(11) | 0.061 | 0(0) | 58(11) | 0.061 | 0(0) | 58(11) | 0.025 |
Given the association between anti-Th/To antibodies and lung disease, we were interested in investigating the frequency of lung cancer in patients with anti-Th/To antibodies. Of the 31 Th/To+ patients with a history of cancer, 7 had lung cancer. Constraining to only the time period within 2, 3, and 5 years surrounding SSc-symptom onset, the number of lung cancer cases were 0, 0, and 0 (all 7 lung cancers were outside of the ±5 year window from SSc-symptom onset). Of the 7 patients, all had radiographic ILD on HRCT imaging, and 6/7 were smokers.
Given the known association of RNApol3 autoantibodies and cancer risk in SSc, we sought to identify whether any patients produced autoantibodies to both RNApol3 and one of the Th/To components, and what the resulting effect on cancer risk these dual immune responses may have. Nine patients produced both hPOP1 and RNApol3 autoantibodies, and none had cancer-associated SSc within 3 years. In contrast, patients with RNApol3 antibodies without antibodies to Th/To components had significantly increased risk of cancer-associated SSc, (24% of SSc patients with RNApol3 autoantibodies have cancer-associated SSc, defined as cancer within 3 years of SSc-onset excluding NMSC, compared to 11% without RNApol3, p<0.0001). Of the 67 patients with antibodies to at least one component of the Th/To complex, few had autoantibodies to the other common SSc specificities: CENP-A (n=2, 3%), CENP-B (n=5, 7%) or Scl-70 (n=5, 7%), as assayed on the Euroimmun Platform. One patient had both CENP-A and CENP-B antibodies.
Discussion
In this manuscript, we describe the prevalence of autoantibodies to four major components of the Th/To complex in SSc patients. This is the largest study performed on Th/To antibodies to date. Our findings highlight the clinical phenotype and negative association of these antibodies with cancer. We reported a prevalence of 8.3% for antibodies against any of the components, with antibody prevalences against the individual complex components ranging from 4 to 9%. This is consistent with prior research of anti-Th/To prevalence defined by immunoprecipitation in cohorts from Japan (4.6%, 14/303) [9], the Netherlands (7%, 12/172 SSc patients with nucleolar pattern on ANA) [20–21], and the United States (15.4%, 23/149) [22]. With regards to the clinical phenotype, we have shown that patients with antibodies to any of the four components are more likely to have limited cutaneous disease and an increased risk of lung involvement. This also is consistent with published findings in SSc cohorts from Japan (36%, 5/14 ILD) [9] and the United States (26%, 6/23 ILD) [22]. In our study, 19–23% of patients with antibodies to one of the four components had PH. This finding is similar to the reported prevalence of PH in different SSc cohorts which showed that 33% (63/193) of anti-Th/To patients have PH [23].
A main aim of this study was to explore the association with cancer for autoantibodies against each of the four Th/To components. Patients with antibodies against hPOP1, RPP25, RPP30, and/or RPP40 were significantly less likely to develop cancer within 2 years of SSc-onset (0% vs 11%, p=0.016). This potential protective effect of anti-hPOP1, -RPP40, -RPP30, and –RPP25 antibodies on cancer-associated SSc is a novel observation. Similar observations have been reported by our group regarding anti-centromere antibodies protecting cancer overall in SSc patients [4]. Given so few patients produced autoantibodies to only 1 or 2 components (the majority produced >2), we were limited in our power to detect any dose-response relationship. Such an observation might support the notion of epitope spreading and warrants further study. Given prior studies demonstrating that the risk of cancer-associated SSc may be modified by the presence of additional immune responses, we were interested in testing whether positivity to any of the anti-Th/To components conferred cancer protection among patients with anti-RNApol3 antibodies. Of the 9 patients who produced both anti-Th/To autoantibodies and anti-RNApol3 antibodies, none had cancer-associated SSc (defined as cancer occurring within 3 years of SSc symptom onset). Although small in number, this observation may signify the ability of anti-Th/To to modify cancer risk conferred by RNApol3 autoantibodies. Similar findings have been reported with the large subunit of RNA polymerase I (RPA194): in a cohort of patients with anti-RNApol3 antibodies, anti-RPA194 was enriched in the group without cancer (18% versus 3.8%, p=0.003) [3]. In this context, it is interesting to note that RNAse P has been reported to play a role in the transcription of RNA polymerases I and III [24–25], and often forms complexes with both enzymes.
The relationship between autoantibodies to individual components of the RNase MRP and RNase P complexes has been poorly described to date. Several prior studies have described the prevalence of anti-Rpp’X’ in small numbers of patients (N<15). Adding to the complexity is that different definitions have been used to describe what constitutes anti-Th/To positivity. In one study [9] of 14 anti-Th/To-positive patients (defined by precipitation of a group of 6 proteins ranging in size from 18–120kDa), 12/14 (86%) were positive for anti-RPP30, and 13/14 (93%) were positive for anti-hPOP1. Another study of 12 anti-Th/To-positive patients (defined as the ability of serum to immunoprecipitate RNase MRP and RNase P RNA) demonstrated that the majority recognized hPOP1 and/or RPP25, but relatively few recognized RPP40 or RPP30 [20]. Given that there are at least 9 individual proteins comprising the Th/To complex, autoantibodies to any of the components could technically be considered positive for Th/To antibodies. Based on the literature, we chose to focus on the four most prevalent autoantigens of the Th/To complex in the current study. A potential limitation of our study is that by not assaying the other components, there is a possible risk of misclassifying the Th/To antibody status of some patients. In the present study, using single-input immunoprecipitation of 35S-methionine-labeled proteins generated by IVTT from the relevant DNAs, the majority of patients (85%) had antibodies to 3 or more components of the Th/To complex. Whether these patients also produce autoantibodies to additional components of the Th/To complex (e.g. hPOP5, RPP14, etc.) warrants further study.
A potential limitation of our study is the possibility that the cancer status may have been misclassified for those patients who developed cancer beyond the 5 year window. Our case-control study design is also a potential limitation, as it provides a cohort enriched for cancer, potentially biasing the general phenotype of SSc patients under study. However, the large number of patients studied likely mitigates this risk. Our cohort did not have comprehensive data on variables to define interstitial lung disease radiographically. As such, we presented data on the presence of a severe restrictive ventilatory defect as a surrogate for ILD. In addition, given the small number of patients with antibodies to only 1 or 2 components of the 4 that were studied, we were unable to comment definitively on differences in demographics or disease characteristics based on different numbers/combinations of anti-Th/To components patients produced. Future studies investigating the order and timing of antibody generation against different components of the Th/To complex are likely to provide additional important insights into how the immune response evolves.
In this study, we assayed autoantibodies to different components of the Th/To complex in a large, meticulously phenotyped cohort of SSc patients with and without cancer. Our data demonstrate that most patients produce autoantibodies to multiple components of the complex, and have a phenotype characterized by pulmonary disease. Our findings show, for the first time, that the presence of any Th/To autoantibody may have a protective effect against contemporaneous cancer. Furthermore, patients producing autoantibodies to the Th/To complex should be vigilantly followed for the development of pulmonary complications of SSc.
Supplementary Material
Acknowledgments
Funding Sources: These studies were supported by the Scleroderma Research Foundation. The Rheumatic Diseases Research Core Center, where the autoantibodies were assayed, is supported by NIH P30-AR070254. LCR and AS are funded in part by the Donald B. and Dorothy L. Stabler Foundation. This work was also supported by NIH grants 1K23AR075898 and AR-073208 from NIAMS. CM is a Jerome L Greene Scholar. BA was supported by T32 AR048522.
The authors report no financial interests that are related to this work.
References
- 1.Shah AA, Rosen A, Hummers L, Wigley F, Casciola-Rosen L. Close temporal relationship between onset of cancer and scleroderma in patients with RNA polymerase I/III antibodies. Arthritis Rheum 2010;62:2787–2795. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shah AA, Xu G, Rosen A, Hummers LK, Wigley FM, Elledge SJ et al. Brief Report: Anti-RNPC-3 Antibodies As a Marker of Cancer-Associated Scleroderma. Arthritis Rheum 2017;69:1306–1312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shah AA, Laiho M, Rosen A, Casciola‐Rosen L. (2019). Scleroderma patients with antibodies against the large subunits of both RNA polymerases‐I and ‐ III are protected against cancer. Arthritis Rheum 2019; 71:1571–1579. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Igusa T, Hummers LK, Visvanathan K, Richardson C, Wigley FM, Casciola-Rosen L et al. Autoantibodies and scleroderma phenotype define subgroups at high-risk and low-risk for cancer. Ann Rheum Dis 2018;77:1179–1186. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Hardin JA, Rahn DR, Shen C, Lerner MR, Wolin SL, Rosa MD et al. Antibodies from patients with connective tissue diseases bind specific subsets of cellular RNA-protein particles. J Clin Invest 1982;70, 141–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Reddy R, Tan EM, Henning D, Nohga K, Busch H. Detection of a nucleolar 7–2 ribonucleoprotein and a cytoplasmic 8–2 ribonucleoprotein with autoantibodies from patients with scleroderma. J Biol Chem 1983;258:1383–1386. [PubMed] [Google Scholar]
- 7.Mahler M, Fritzler MJ, Satoh M. Autoantibodies to the mitochondrial RNA processing (MRP) complex also known as Th/To autoantigen. Autoimmun Rev 2015;14:254–257. [DOI] [PubMed] [Google Scholar]
- 8.Mattijssen S, Welting TJM, Pruijn GJM. RNase MRP and disease. Wiley Interdiscip Rev RNA. 2010;1:102–116. [DOI] [PubMed] [Google Scholar]
- 9.Kuwana M, Kimura K, Hirakata M, Kawakami Y, Ikeda Y. Differences in autoantibody response to Th/To between systemic sclerosis and other autoimmune diseases. Ann Rheum Dis 2002;61:842–846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Mahler M, Satoh M, Hudson M, Baron M, Chan JYF, Chan EK et al. Autoantibodies to the Rpp25 Component of the Th/To Complex are the Most Common Antibodies in Patients with Systemic Sclerosis without Antibodies Detectable by Widely Available Commercial Tests. J Rheum 2014;41:1334–1343. [DOI] [PubMed] [Google Scholar]
- 11.Mitri GM, Lucas M, Fertig N, Steen VD, Medsger TA. A comparison between anti-Th/To- and anticentromere antibody-positive systemic sclerosis patients with limited cutaneous involvement. Arthritis Rheum 2003;48:203–209. [DOI] [PubMed] [Google Scholar]
- 12.Shah AA, Casciola-Rosen L. Mechanistic and clinical insights at the scleroderma-cancer interface. J Scleroderma Relat Disord 2017;2:153–159. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Medsger TA, Bombardieri S, Czirjak L, Scorza R, Della Rossa A, Bencivelli W. Assessment of disease severity and prognosis. Clin Exp Rheumatol 2003;21:S42–6. [PubMed] [Google Scholar]
- 14.Hankinson JL, Odencrantz JR, Fedan KB. Spirometric reference values from a sample of the general U.S. population. Am J Respir Crit Care Med 1999;159:179–187. [DOI] [PubMed] [Google Scholar]
- 15.Knudson RJ, Kaltenborn WT, Knudson DE, Burrows B. The single-breath carbon monoxide diffusing capacity. Reference equations derived from a healthy nonsmoking population and effects of hematocrit. Am Rev Respir Dis 1987;135:805–811. [DOI] [PubMed] [Google Scholar]
- 16.Cottrell TR, Wise RA, Wigley FM, Boin F. The degree of skin involvement identifies distinct lung disease outcomes and survival in systemic sclerosis. Ann Rheum Dis 2014;73:1060–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mecoli CA, Shah AA, Boin F, Wigley FM, Hummers LK. Vascular complications in systemic sclerosis: a prospective cohort study. Clin Rheumatol 2018;37:2429–2437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Simonneau G, Montani D, Celermajer DS, Denton CP, Gatzoulis MA, Krowka M et al. Haemodynamic definitions and updated clinical classification of pulmonary hypertension. Eur Respir J. 2019;24:53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Fiorentino D, Chung L, Zwerner J, Rosen A, Casciola-Rosen L. The mucocutaneous and systemic phenotype of dermatomyositis patients with antibodies to MDA5 (CADM-140): a retrospective study. J Am Acad Dermatol 2011;65:25–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Van Eenennaam H, Vogelzangs JHP, Lugtenberg D, Van Den Hoogen FHJ, Van Venrooij WJ, Pruijn GJM. Identity of the RNase MRP- and RNase P-associated Th/To autoantigen. Arthritis Rheum 2002;46:3266–3272. [DOI] [PubMed] [Google Scholar]
- 21.Van Eenennaam H, Nayef Jarrous, Walth H, Jarrous N, van Venrooij WJ, Pruijn GJ. Architecture and Function of the Human Endonucleases RNase P and RNase MRP. IUBMB Life 2000;49:265–272. [DOI] [PubMed] [Google Scholar]
- 22.Falkner D, Wilson J, Medsger TA, Morel PA. HLA and clinical associations in systemic sclerosis patients with anti-Th/To antibodies. Arthritis Rheum 1998;41:74–80. [DOI] [PubMed] [Google Scholar]
- 23.Nunes JPL, Cunha AC, Meirinhos T, Nunes A, Araújo PM, Godinho AR, et al. Prevalence of auto-antibodies associated to pulmonary arterial hypertension in scleroderma – A review. Autoimmun Rev 2018;17:1186–1201. [DOI] [PubMed] [Google Scholar]
- 24.Reiner R, Ben-Asouli Y, Krilovetzky I, Jarrous N. A role for the catalytic ribonucleoprotein RNase P in RNA polymerase III transcription. Genes Dev 2006;20:1621–1635. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Reiner R, Krasnov-Yoeli N, Dehtiar Y, Jarrous N. Function and assembly of a chromatin-associated RNase P that is required for efficient transcription by RNA polymerase I. PloS One 2008;3:e4072. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.