Abstract
Both metachronous and synchronous tumors pose a diagnostic and clinical challenge, more so when one of the specimens demonstrates the rare neuroendocrine histology. We describe a patient with sarcoidosis who was treated for endometrial and ovarian neoplasm, recurred with two separate histologies (adenocarcinoma and high grade neuroendocrine), both associated with microsatellite instability (MSI)‐high status. Targeted next‐generation sequencing of tumor with synonymous somatic alterations pointed to a common ancestry of all three tumors and patient was successfully treated with a tailored immunotherapy regimen. Her sarcoidosis worsened only slightly, and immunotherapy did not need to be discontinued. This case highlights the importance of molecular testing for the optimal therapy of complex synchronous tumors and the need for communication between surgical and medical oncologists in patients with MSI‐high cancer.
Key Points
The case of a patient with a recurrent gynecological cancer presenting as microsatellite instability (MSI)‐high endometrial adenocarcinoma and MSI‐high neuroendocrine tumor is reported.
This case demonstrated a common genetic lineage with good response to checkpoint inhibition without clinical worsening of autoimmune disease.
This article adds to the literature, suggesting tumor evolution with neuroendocrine differentiation in some cancers, and argues that a molecular‐based approach to treatment might achieve better understanding and possibly superior treatment outcomes.
Short abstract
This case report highlights the importance of molecular testing for optimal therapy of complex synchronous tumors and the need for communication between surgical and medical oncologists in cases of high microsatellite instability cancer.
Patient Story
Neuroendocrine neoplasms (NENs) of the cervix and endometrium are uncommon, constituting a mere 2% of gynecologic cancers [1]. On some occasions, NENs can coexist with another histology, usually of the adenocarcinoma type. These synchronous tumors can form separate components or present in the same biopsy specimen, in which case they sometimes are categorized as mixed neuroendocrine and non‐neuroendocrine (MiNEN) neoplasms. Synchronous tumors are particularly challenging because of their unexpected behaviors and require an individualized approach based on perceived relative aggressiveness of their parts. Their origin is also an area of active debate: given significant morphological and immunohistochemical differences in the separate histologies, it is unknown if they represent synchronous malignancies or tumor transformation from one phenotype to the other. With the advent of next‐generation sequencing, this question can be addressed at the molecular level and sometimes allows for tailored and personalized treatment for all histologies. In this article, we have tracked the evolution of a recurrent ovarian and endometrial adenocarcinoma into a phenotypically distinct but genetically similar neoplasm, at a site of distant metastasis.
A 45‐year‐old female patient was diagnosed with endometrial adenocarcinoma (FIGO1 T1aN0M0 stage Ia) and synchronous ovarian endometrioid adenocarcinoma (FIGO1 T1cN0M0 stage Ic), with similar pathology and tumor cellularity of 80%. She underwent curative resection, followed by adjuvant carboplatin and paclitaxel. Mismatch repair protein deficiency testing on the tumor showed loss of MSH2 and MSH6. Her mother had cervical cancer, but she had a complete genetic evaluation with a 32‐gene panel that showed no germline mutations.
She remained disease free for about 6 years, with interim diagnosis of hepatosplenic sarcoidosis, which was treated with methotrexate and azathioprine. She then complained of abdominal pain, and computed tomography (CT) demonstrated an enhancing, lobulated mass within the right hemipelvis, as well as hypoenhancing liver lesions superimposed on hepatic sarcoidosis (Fig. 1).
Figure 1.
Computed tomography of the abdomen on disease recurrence. (A): Axial contrast‐enhanced computed tomography demonstrated an enhancing lobulated mass (*) in the right hemipelvis, corresponding to biopsy proven endometroid carcinoma. (B): More importantly, there is a hypoenhancing mass (arrowheads) within segment 6 of the liver corresponding to biopsy proven neuroendocrine tumor; other similar appearing lesions were present elsewhere in the liver (not shown). The tiny hypodensities in the background liver correspond to hepatic involvement with sarcoidosis.
Molecular Tumor Board
Both the pelvic mass and liver lesion were biopsied. Tumor cellularity was high, at 70% on liver and pelvic mass. Pathology was morphologically and immunohistochemically distinct between the pelvic mass and the sampled liver lesion. The pelvic mass was consistent with recurrent endometrioid carcinoma, with strong estrogen receptor (ER) staining. The liver lesion showed nests of relatively monomorphic neoplastic cells with rounded nuclei, granular chromatin, and moderate amount of eosinophilic and amphophilic cytoplasm and without gland formation, significant nuclear atypia, or necrosis. The liver neoplasm stained positively for chromogranin and synaptophysin and negatively for ER and had a Ki‐67 proliferative index of 40%, and a diagnosis of well‐differentiated neuroendocrine tumor, World Health Organization (WHO) grade 3 was made (Fig. 2; Table 1). There were no tumor‐infiltrating lymphocytes (TILs) on any of the biopsy specimens.
Figure 2.
H&E on original ovarian and endometrial endometrioid adenocarcinoma. Ovarian (A); endometrial (B). Both had a similar morphologic appearance and were positive for ER and progesterone receptor (PR) and negative for chromogranin and synaptophysin. (C): Recurrent pelvic tumor shows a gland forming neoplasm with more complexity but otherwise similar appearance, ER positivity, and negative synaptophysin staining. (D): Liver biopsy shows nests of monomorphic cells with rounded nuclei, granular chromatin, and moderate amount of eosinophilic and amphophilic cytoplasm; the result was positive for synaptophysin and only weakly and focally positive for ER.Abbreviation: ER, estrogen receptor.
Table 1.
Immunohistochemistry of the three tumor specimens
| Specimen | Diagnosis | CK7 | CK20 | CDX2 | PAX8 | ER | PR | Synaptophysin | Chromogranin |
|---|---|---|---|---|---|---|---|---|---|
| Ovary and uterus | Endometrial carcinoma | Minority + | − | Weak, focal + | − | + | + | − | − |
| Liver tumor | Well‐differentiated NET G3 | − | − | Strong, patchy + | − | Focal + | − | Majority + | Patchy + |
| Pelvic mass | Metastatic endometrioid carcinoma | Rare + | − | Scattered + | − | + | N/A | − | N/A |
Abbreviations: ER, estrogen receptor; N/A, not applicable; NET, neuroendocrine tumor; PR, progesterone receptor; −, negative; +, positive.
Commercially available targeted next‐generation sequencing (NGS; FoundationOne) on both the pelvic and liver lesions, as well as the primary endometrioid adenocarcinoma involving the ovary, showed a microsatellite instability (MSI)‐high status, high tumor mutational burden and striking similarity between the original and recurrent tumor specimens, both in established variants as well as variants of unknown significance (Table 2). Programmed death‐ligand 1 (PD‐L1) expression increased from 0% on the original tumor to 5% in the adenocarcinoma and 30% in the NEN recurrence. Loss of MSH2 and MSH6 was seen in the original hysterectomy and oophorectomy specimen and the recurrent pelvic mass, whereas only MSH2 loss was seen in the liver lesion demonstrating the neuroendocrine tumor. In an attempt to further clarify the origin of the tumors, we examined the synonymous somatic alterations (supplemental online Appendix 1). Two hundred and seventy‐five out of a total of 292 (94%) were shared among three tumor types.
Table 2.
Next generation sequencing with VAF for known variants, as well as VUS
| Specimen (all MSI‐high) | Original ovarian and uterine tumor | Adenocarcinoma recurrence | NEN recurrence |
|---|---|---|---|
| VAF | |||
| PD‐L1 expression (%) | 0 | 5 | 30 |
| TMB | 32 | 44 | 44 |
| APC R232* | 36.4 | 29.1 | 30.2 |
| APC splice site 423‐2A>G | 38.2 | 31.7 | 30.2 |
| APC T1556fs*9 | 32.4 | 23.4 | 23.2 |
| ARID1A G276fs*87 | 33.8 | 30.9 | 24.2 |
| ARID1A Q1327fs*11 | 24.3 | ||
| BCORL1 P1681fs*20 | 35.5 | 25.3 | 26 |
| ERBB2 R678Q | 39.9 | 25.3 | 25.9 |
| ESR1 Y52H | 16.4 | ||
| FBXW7 R13* | 37.2 | 27.3 | 26.8 |
| FUBP1 Y505fs*20 | 65.3 | 50.5 | 48.3 |
| MLL2 P647fs*283 | 24 | ||
| MSH2 splice site 942 + 1G>T | 38.8 | 26.7 | 27.4 |
| MSH2 C778fs*9 | 34.4 | 26.9 | 23 |
| MSH6 F1088fs*5 | 28.5 | 19.8 | |
| PBRM1 G677fs*9 | 34.9 | 28.7 | 24.5 |
| PBRM1 L1349fs*35 | 27.4 | ||
| PPP2R1A R258C | 39 | 27.9 | 25.9 |
| PRKAR1A R368* | 26 | ||
| PTEN N323fs*21 | 30.9 | 27.5 | 26.9 |
| PTEN R233* | 36.4 | 27.5 | 28.2 |
| QKI K134fs*14 | 33.7 | 23.3 | 22.9 |
| SDHA Q176H | 3 | ||
| JAK1 K860fs*16 | 24.3 | ||
| VHL F136V | 1.4 | ||
| SOX9 V306fs*77 | 8.7 | ||
| VUS | |||
| ACVR1B R530H | + | ||
| ALOX12B F211del | + | + | + |
| APC R640Q | + | ||
| ARID1A R811S | + | ||
| ATM W2491C | + | + | + |
| ATR T1989A | + | + | |
| AXL Q361P | + | + | + |
| BRAF R178* | + | + | + |
| BRCA2 R1217K | + | + | + |
| CARD11 R555fs*45 | + | + | |
| CBFB F18fs*4 | + | + | |
| CYP17A1 D289Y | + | ||
| DAXX R291W | + | + | + |
| DDR1 R93* | + | + | + |
| DDR1 W385fs*75 | + | ||
| EP300 V2340I | + | ||
| DOT1L A1448V | + | ||
| EP300 A488T and P866H | + | ||
| EPHA3 E647fs*9 | + | + | + |
| IKZF1 D290G | + | + | + |
| JAK2 P1002S | + | + | + |
| KDM6A C468R | + | + | + |
| MED12 C982Y | + | ||
| MEF2B I6fs*14 | + | + | + |
| MEN1 R340W | + | + | |
| NBN K156N | + | ||
| NOTCH1 A2019T | + | + | + |
| PDCD1 (PD‐1) V43M | + | + | + |
| PIK3CA R852Q | + | + | + |
| POLD1 D987fs*58 | + | + | + |
| POLE G1343S | + | + | + |
| PTPRO R1082W | + | + | + |
| RET L56M and P996Q | + | + | + |
| SDHA L204S | + | ||
| SMARCA4 V855I | + | + | + |
| SMARCB1 K364del | + | + | + |
| SRC R271Q | + | ||
| SYK A7T | + | + | + |
| TSC1 R284C | + |
Abbreviations: +, present; MSI, microsatellite instability; NEN, neuroendocrine neoplasm; PD‐L1, programmed death‐ligand 1; TMB, tumor mutational burden; VAF, variant allele frequency; VUS, variants of unknown significance.
She received chemotherapy with carboplatin and etoposide after coordination with the gynecological oncology service. She responded briefly but progressed after 5 cycles. Given MSI‐high status, she was switched to immunotherapy with pembrolizumab, with favorable (partial) response in both the adenocarcinoma and neuroendocrine components (Fig. 3) on first CT, and ongoing response more than 1 year later. Her sarcoidosis worsened slightly with the initiation of immunotherapy, but she was observed without symptoms with input from rheumatology. On restaging imaging, her sarcoidosis lesions were stable, her tumor continued to shrink in all locations, and she remained asymptomatic.
Figure 3.
Eight months after initiation of pembrolizumab, contrast‐enhanced computed tomography images through similar levels as Figure 1 show marked decrease in size of pelvic mass as well as liver mass. Pelvic mass (A), shown by asterisk (*); liver mass (B), shown by arrowheads. Other liver metastases similarly demonstrated decrease in size (not shown).
Discussion
In this article, we describe a patient with no familial genetic syndrome who developed three instances of an MSI‐high tumor: Endometrial and ovary status post curative treatment, then, much later, endometrial recurrence and liver metastasis. The liver metastasis was found to be a grade III well‐differentiated neuroendocrine tumor. However, genetic analysis suggested common origin of all samples, and the patient ultimately received U.S. Food and Drug Administration (FDA)‐approved immunotherapy [2], with excellent durable results and no significant worsening of her autoimmune disease.
We have verified that the original endometrial and ovarian tumor had similar histologies, and the same histology was reflected on the recurrence in the endometrium many years later. We also extensively retrospectively stained the samples to exclude a mixed neuroendocrine neoplasm. The recurrent liver metastasis showed focal, weak nuclear positivity for ER but strong expression of synaptophysin. No other samples expressed neuroendocrine markers. There were no observable TILs. Despite their differences, tumors were strikingly similar on a molecular level. NGS on all three specimens was suggestive of a common lineage with multiple shared mutations (both deleterious, as well as of unknown significance), and all tumors remained MSI‐high. Moreover, somatic variant testing was strikingly similar between all three instances of the tumor (94% common variants between all cancer instances), strengthening the suspicion of a common origin.
Although rare, the occurrence of two or more synchronous neoplasms has been well described in the literature and was considered in our case. Multiple cancers can arise in the setting of permissive germline disorders such as familial syndromes. Lynch syndrome and mismatch repair deficiency [3, 4], which allows for accumulation of DNA replication errors over time, is one of these. Our patient had a family history of cervical cancer (mother) but did not have a germline mutation by commercial panel that included BRCA, MMR, POLE, and PTEN genes (full gene list in supplemental online Appendix). Moreover, although some driver mutations can be common in patients with familial syndromes, the degree of similarity of NGS and somatic variants argues against three randomly created neoplasms. Ideally, this would be confirmed with whole‐genome sequencing, but this was not available for our patient.
Another point is that we are not sure about the homogeneity of the liver tumor metastasis, as the specimen showed focal, weak nuclear positivity for ER. Whether the neoplasm in the liver entirely represents the neuroendocrine component or presents as a MiNEN cannot be determined, because this sampling was only from a needle biopsy. MiNENs are known to be very heterogeneous and are composed of an intermixed population of different histologies, with a randomly assigned percentage of each component amounting to at least 30% of the entire neoplasm. Collision tumors (two separate histologies that only share a border but have different cellular origins) and amphicrine tumors (where the same cells express variable lineage‐specific phenotypical markers) are not included in the category of MiNENs (WHO 2019: Tumours of the Digestive Tract).
It is also worth noting that transformation to a neuroendocrine phenotype has been described in certain neoplasms, although the underlying process is not very clear. For example, lineage plasticity has been demonstrated in lung cancer after treatment with EGFR‐targeting agents [5], and long‐term androgen deprivation in prostate cancer can lead to treatment‐emergent neuroendocrine tumors, usually by loss of RB/p53 function and decreased activity of SOX2 or EZH2 [6, 7]. In gastrointestinal [8] and lung [9] MiNENs, genetic and immunohistochemical studies have suggested a common origin in the amphicrine but not other subtypes. In our case, a crude comparison of significant variants that appeared in the neuroendocrine component shows emerging JAK1, PRKAR1A, and SOX9 mutations, some of which have been loosely associated with a neuroendocrine phenotype in early studies of multiple tumors but could also be tissue specific or random events [10, 11, 12, 13]. It would take comparisons with multiple other similar tumors and considerations of the variants that disappeared in evolution to reach definite conclusions.
Our patient benefitted from FDA‐approved checkpoint inhibition [2] with long‐lasting results (ongoing response more than a year later), probably attributed to her high mutational burden. Lynch syndrome accounts for less than 5% of all endometrial carcinomas in large studies [14], but MMR deficiency has been described in about 44% of endometrial MiNENs in a small series of patients [15], with all cases being mixed histologies. Sharabi et al. [16] reported a cervical neuroendocrine carcinoma with MSI‐high status that responded to stereotactic body radiation therapy and nivolumab and argued for an abscopic effect of radiotherapy. In that paper, the reported molecular alterations shared some similarities with our case (mainly FBXW7, MSH2, PTEN, QKI, and JAK1 mutations), but there was no mixed histology or evidence of a separate primary as mentioned above. With the approval tumor agnostic therapies, such as those targeting the TRK [17] fusion and possibly others targeting RET or FGFR, it is becoming increasingly important to know the mutational backbone of tumors. Although this might not be feasible or cost effective in every case, it definitely should be considered in instances like ours, where multiple, grossly different histologies are involved.
One of the complicating factors in this case is that the patient was diagnosed with systemic (spleen, lung, liver) granulomatous noncaseating sarcoidosis from a pelvic lymph node biopsy. This was years after her initial treatment and before her recurrence was established, which is an interesting phenomenon. She was treated with methotrexate followed by azathioprine (stopped with the diagnosis of malignancy). Inflammatory autoimmune disease has been associated with a higher incidence of cancer [18], with sarcoidosis in one study mainly linked to gastrointestinal and skin malignancies. Chronic use of immunosuppressants such as azathioprine has been associated with higher risk of secondary malignancy in patients with inflammatory bowel disease [19, 20], but we feel that the patient developed a recurrence and not a new malignancy. The continuous immunosuppressive effect might have accelerated the new growth, however.
With pembrolizumab treatment, the patient's chest imaging worsened only temporarily, and pulmonary function tests remained unaffected. There is no concrete evidence on the safety of continuing immunotherapy with checkpoint inhibitors in patients with established sarcoidosis, so this decision was made on an institutional level after talking to rheumatology. Sarcoidosis‐like reactions have been described after checkpoint inhibitor therapy [21], and patients with sarcoidosis and cancer have been safely treated with immunotherapy combinations [22, 23] with salvage immunosuppression if needed. Treatment of patients with non‐life threatening autoimmune disease on low level of immunosuppression or no immunosuppression is in accordance with the National Comprehensive Cancer Network recommendations [24]. Finally, there is no known association between sarcoidosis and checkpoint inhibitor efficacy to our knowledge. Most reports addressing immunotherapy and autoimmune disease include patients who were responding to immunotherapy, but there is an obvious selection bias, as these patients were treated long enough to observe worsening of the immune disease.
Patient Update
Given asymptomatic sarcoidosis status, and after a lot of discussions with rheumatology and the patient, we chose to treat with close monitoring. Sarcoidosis lesions worsened slightly on initial scan, then remained stable on subsequent scans, and the patient remains generally asymptomatic and without progression more than a year later.
Author Contributions
Conception/design: Nikolaos A. Trikalinos
Provision of study material or patients: Nikolaos A. Trikalinos, Deyali Chatterjee, Motoyo Yano
Collection and assembly of data: Nikolaos A. Trikalinos, Deyali Chatterjee, Motoyo Yano
Data analysis and interpretation: Nikolaos A. Trikalinos, Deyali Chatterjee, Kyle Winter, Matthew Powell, Motoyo Yano
Manuscript writing: Nikolaos A. Trikalinos, Deyali Chatterjee, MY
Final approval of manuscript: Nikolaos A. Trikalinos, Deyali Chatterjee, Kyle Winter, Matthew Powell, Motoyo Yano
Accountable for all aspects of the work: Nikolaos A. Trikalinos, Deyali Chatterjee, Kyle Winter, Matthew Powell, Motoyo Yano
Disclosures
Matthew Powell: Tesaro/GlaxoSmithKline, AstraZeneca, Merck, Clovis Oncology, Eisai (C/A). The other authors indicated no financial relationships.
(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/inventor/patent holder; (SAB) Scientific advisory board
Supporting information
See http://www.TheOncologist.com for supplemental material available online.
Appendix S1: Supplementary Information
Acknowledgments
Authors have obtained informed consent from the patient for publication of the submitted article and accompanying images.
Disclosures of potential conflicts of interest may be found at the end of this article.
No part of this article may be reproduced, stored, or transmitted in any form or for any means without the prior permission in writing from the copyright holder. For information on purchasing reprints contact Commercialreprints@wiley.com. For permission information contact permissions@wiley.com.
References
- 1. Rouzbahman M, Clarke B. Neuroendocrine tumors of the gynecologic tract: Select topics. Semin Diagn Pathol 2013;30:224–233. [DOI] [PubMed] [Google Scholar]
- 2. Le DT, Uram JN, Wang H et al. PD‐1 blockade in tumors with mismatch‐repair deficiency. N Engl J Med 2015;372:2509–2520. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Dowty JG, Win AK, Buchanan DD et al. Cancer risks for MLH1 and MSH2 mutation carriers. Hum Mutat 2013;34:490–497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Antill YC, Dowty JG, Win AK et al. Lynch syndrome and cervical cancer. Int J Cancer 2015;137:2757–2761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Yuan S, Norgard RJ, Stanger BZ. Cellular plasticity in cancer. Cancer Discov 2019;9:837–851. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sharma A, Yeow WS, Ertel A et al. The retinoblastoma tumor suppressor controls androgen signaling and human prostate cancer progression. J Clin Invest 2010;120:4478–4492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Mu P, Zhang Z, Benelli M et al. SOX2 promotes lineage plasticity and antiandrogen resistance in TP53‐ and RB1‐deficient prostate cancer. Science 2017;355:84–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. de Mestier L, Cros J, Neuzillet C et al. Digestive system mixed neuroendocrine‐non‐neuroendocrine neoplasms. Neuroendocrinology 2017;105:412–425. [DOI] [PubMed] [Google Scholar]
- 9. Iijima M, Yokobori T, Mogi A et al. Genetic and immunohistochemical studies investigating the histogenesis of neuroendocrine and carcinomatous components of combined neuroendocrine carcinoma. Ann Surg Oncol 2019;26:1744–1750. [DOI] [PubMed] [Google Scholar]
- 10. Danziger O, Pupko T, Bacharach E et al. Interleukin‐6 and interferon‐α signaling via JAK1‐STAT differentially regulate oncolytic versus cytoprotective antiviral states. Front Immunol 2018;9:94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Gu L, Talati P, Vogiatzi P et al. Pharmacologic suppression of JAK1/2 by JAK1/2 inhibitor AZD1480 potently inhibits IL‐6‐induced experimental prostate cancer metastases formation. Mol Cancer Ther 2014;13:1246–1258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Saloustros E, Salpea P, Starost M et al. Prkar1a gene knockout in the pancreas leads to neuroendocrine tumorigenesis. Endocr Relat Cancer 2017;24:31–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Lee CJ, Sung PL, Kuo MH et al. Crosstalk between SOX2 and cytokine signaling in endometrial carcinoma. Sci Rep 2018;8:17550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kwon JS, Scott JL, Gilks CB et al. Testing women with endometrial cancer to detect Lynch syndrome. J Clin Oncol 2011;29:2247–2252. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Pocrnich CE, Ramalingam P, Euscher ED et al. Neuroendocrine carcinoma of the endometrium: A clinicopathologic study of 25 cases. Am J Surg Pathol 2016;40:577–586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Sharabi A, Kim SS, Kato S et al. Exceptional response to nivolumab and stereotactic body radiation therapy (SBRT) in neuroendocrine cervical carcinoma with high tumor mutational burden: Management considerations from the Center For Personalized Cancer Therapy at UC San Diego Moores Cancer Center. The Oncologist 2017;22:631–637. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Drilon A, Laetsch TW, Kummar S et al. Efficacy of larotrectinib in TRK fusion–positive cancers in adults and children. N Engl J Med 2018;378:731–739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Franks AL, Slansky JE. Multiple associations between a broad spectrum of autoimmune diseases, chronic inflammatory diseases and cancer. Anticancer Res 2012;32:1119–1136. [PMC free article] [PubMed] [Google Scholar]
- 19. Pasternak B, Svanström H, Schmiegelow K et al. Use of azathioprine and the risk of cancer in inflammatory bowel disease. Am J Epidemiol 2013;177:1296–1305. [DOI] [PubMed] [Google Scholar]
- 20. Fraser AG, Orchard TR, Robinson EM et al. Long‐term risk of malignancy after treatment of inflammatory bowel disease with azathioprine. Aliment Pharmacol Ther 2002;16:1225–1232. [DOI] [PubMed] [Google Scholar]
- 21. Gkiozos I, Kopitopoulou A, Kalkanis A et al. Sarcoidosis‐like reactions induced by checkpoint inhibitors. J Thorac Oncol 2018;13:1076–1082. [DOI] [PubMed] [Google Scholar]
- 22. Johnson DB, Sullivan RJ, Ott PA et al. Ipilimumab therapy in patients with advanced melanoma and preexisting autoimmune disorders. JAMA Oncol 2016;2:234–240. [DOI] [PubMed] [Google Scholar]
- 23. Menzies AM, Johnson DB, Ramanujam S et al. Anti‐PD‐1 therapy in patients with advanced melanoma and preexisting autoimmune disorders or major toxicity with ipilimumab. Ann Oncol 2017;28:368–376. [DOI] [PubMed] [Google Scholar]
- 24. Kennedy LC, Bhatia S, Thompson JA et al. Preexisting autoimmune disease: Implications for immune checkpoint inhibitor therapy in solid tumors. J Natl Compr Canc Netw 2019;17:750–757. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
See http://www.TheOncologist.com for supplemental material available online.
Appendix S1: Supplementary Information