Sjogren's syndrome due to immune checkpoint inhibitors (ICIs): Insights from a single-institution series and systematic review of the literature

Case 1

A 79-year-old Caucasian man had history of surgically resected early-stage malignant melanoma of the skin of the mid-lower back in 1980. In the summer of 2019, he noticed a small subcutaneous nodule in the medial aspect of the left elbow that gradually increased in size. He had excision of the nodule and sentinel lymph node biopsy in December 2019; pathology showed metastatic malignant melanoma and the sentinel node was negative. There were 6–10 mitoses per high-power field; BRAF mutation was negative. On thorough skin examination, no primary lesion was identified. A positron emission tomography–computed tomography (PET/CT) scan from the vertex to the toes and a magnetic resonance imaging (MRI) of the brain were negative for any other sites of disease. In July 2020, adjuvant nivolumab was started at 480 mg IV every 4 weeks. After the first dose, he developed bilateral shoulder myalgias and weakness, elevated erythrocyte sedimentation rate (ESR) and was diagnosed with PMR. The PMR responded well to oral prednisone, 10 mg daily.

In December 2020, he developed dry mouth with difficulty chewing food and dry eyes with occasional redness and irritation. His symptoms were mild on presentation, but progressed over the next four months, despite using artificial tears, artificial saliva and lemon juice for stimulating saliva production. Physical examination revealed geographical tongue with dry oral mucosa and conjunctival erythema. There was no pooling of saliva in the floor of the mouth yielding a score of 4 on the Challacombe scale of clinical oral dryness. Schirmer test was positive (filter paper remained less than 5 mm of wetting after 5 min of continuous contact). Antinuclear antibodies (ANA), hepatitis C virus (HCV) antibodies, human immunodeficiency virus (HIV) antibodies, rheumatoid factor (RF), anti-Sjogren's syndrome type A antibodies (SS-A) and anti-Sjogren's syndrome type B antibodies (SS-B) all were negative. The patient declined buccal mucosal biopsy, but continued using artificial tears and saliva. He was able to continue the nivolumab regimen, without dose reduction or interruption. In June 2022, he developed bilateral lung infiltrates which prompted discontinuation of nivolumab. A follow up visit at six weeks revealed improvement in his dry eye and dry mouth symptoms, and he reported no need for further use of artificial tears or saliva. At 6-month follow up, the patient had normal lacrimation and salivation. The calculated cause-effect probability using the Naranjo nomogram showed the relationship between nivolumab and the patient symptoms to be probable (Table 1).

OPEN IN VIEWER

Table 1.

The Naranjo Adverse Drug Reaction (ADR) probability scale questionnaire in the patient #1.

To assess the adverse drug reaction, please answer the following questionnaire and give the pertinent score.	Yes	No	Do not know	Score
1. Are there previous conclusive reports on this reaction?	+1	0	0	+1
2. Did the adverse event appear after the suspected drug was administered?	+2	−1	0	+2
3. Did the adverse event improve when the drug was discontinued or a specific antagonist was administered?	+1	0	0	+1
4. Did the adverse event reappear when the drug was readministered?	+2	−1	0	0
5. Are there alternative causes that could on their own have caused the reaction?	−1	+2	0	+2
6. Did the reaction reappear when a placebo was given?	−1	+1	0	0
7. Was the drug detected in the blood or other fluids in concentrations known to be toxic?	+1	0	0	0
8. Was the reaction more severe when the dose was increased or less severe when the dose was decreased?	+1	0	0	0
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0
10. Was the adverse event confirmed by any objective evidence?	+1	0	0	+1
Score: 7

The Naranjo criteria classify the probability that an adverse event is related to drug therapy based on a list of weighted questions, which examine factors such as the temporal association of drug administration and event occurrence, alternative causes for the event, drug levels, dose-response relationships and previous patient experience with the medication. The ADR is assigned to a probability category from the total score as follows: definite if the overall score is 9 or greater, probable for a score of 5–8, possible for 1–4, and doubtful if the score is 0. The Naranjo criteria do not take into account drug-drug interactions. Drugs are evaluated individually for causality, and points are deducted if another factor may have resulted in the adverse event, thereby weakening the causal association.

Case 2

A 76-year-old Caucasian man was originally diagnosed with intermediate-risk malignant melanoma of the skin of the right medial lower leg in 2015. The tumor was surgically excised; Breslow thickness was 1.5 mm, and sentinel lymph node was negative. He subsequently had two other early-stage skin melanomas; one in the umbilical area and the other in the midline of the upper back, both of which were excised in June 2016.

In September 2020, he developed another 1.4 cm lesion in the medial aspect of the right thigh for which he underwent radical resection and regional node biopsy. Final pathology showed a negative node, but the lesion in the thigh measured 1.5 × 1.4 × 1.3 cm consistent with recurrent/metastatic melanoma. BRAF mutation was negative. The primary lesion was thought to originate in the skin of right calf in 2015, and he had a subcutaneous metastasis to the right thigh. In November 2021, adjuvant nivolumab was started at 400 mg IV every 4 weeks. After the 2^nd dose, the patient reported of a “sandpaper” sensation in the buccal area and a foreign body sensation in both eyes. He also developed a grade 1 maculo-papular rash. He soon developed severe dry mouth causing problems with chewing food, despite artificial saliva use before meals. He had persistent dry eyes, despite using artificial eye drops regularly. On examination, bilateral conjunctivae and inner eyelids were erythematous. The oral mucosa had a shiny appearance with smooth gingiva with no pooling of saliva suggesting a score of 6 on the Challacombe scale of clinical oral dryness. The unstimulated whole salivary flow rate was about 0.1 ml/min. The Schirmer test was positive (filter paper remained less than 3 mm of wetting after 5 min of continuous contact). ANA, RF, HIV, HCV, SS-A and SS-B antibodies were negative.

He continued to use artificial eye drops, and artificial saliva, and was able to tolerate the nivolumab regimen for a total of 12 months. In September 2022, he developed modestly elevated liver enzymes and mild hypothyroidism, in addition to the non-resolving Sjogren's syndrome. After nivolumab discontinuation, the patient reported some improvement of his dry eyes and mouth. After 6 months, his dry eye symptoms resolved but the dry mouth persisted and he continued to use artificial saliva and lemon juice. The calculated cause-effect probability using the Naranjo nomogram showed the relationship between nivolumab and the patient symptoms to be probable (Table 2).

OPEN IN VIEWER

Table 2.

The Naranjo Adverse Drug Reaction (ADR) probability scale questionnaire in the patient #2.

To assess the adverse drug reaction, please answer the following questionnaire and give the pertinent score.	Yes	No	Do not know	Score
1. Are there previous conclusive reports on this reaction?	+1	0	0	+1
2. Did the adverse event appear after the suspected drug was administered?	+2	−1	0	+2
3. Did the adverse event improve when the drug was discontinued or a specific antagonist was administered?	+1	0	0	1+
4. Did the adverse event reappear when the drug was readministered?	+2	−1	0	0
5. Are there alternative causes that could on their own have caused the reaction?	−1	+2	0	+2
6. Did the reaction reappear when a placebo was given?	−1	+1	0	0
7. Was the drug detected in the blood or other fluids in concentrations known to be toxic?	+1	0	0	0
8. Was the reaction more severe when the dose was increased or less severe when the dose was decreased?	+1	0	0	0
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0
10. Was the adverse event confirmed by any objective evidence?	+1	0	0	+1
Score: 7

The Naranjo criteria classify the probability that an adverse event is related to drug therapy based on a list of weighted questions, which examine factors such as the temporal association of drug administration and event occurrence, alternative causes for the event, drug levels, dose-response relationships and previous patient experience with the medication. The ADR is assigned to a probability category from the total score as follows: definite if the overall score is 9 or greater, probable for a score of 5–8, possible for 1–4, and doubtful if the score is 0. The Naranjo criteria do not take into account drug-drug interactions. Drugs are evaluated individually for causality, and points are deducted if another factor may have resulted in the adverse event, thereby weakening the causal association.

Case 3

A 78-year-old Caucasian man, with a history of early-stage bladder cancer diagnosed in March 2014, was initially managed with transurethral resection of bladder tumor (TURBT) followed by mitomycin C bladder instillations. In August 2018, a CT scan showed a 3 cm right middle lobe lung mass and biopsy showed metastatic urothelial cancer. He received six cycles of cisplatin-gemcitabine from September 2018 to February 2019.

In December 2018, PET/CT scan showed a decrease in size and hypermetabolism of the lung mass. In July 2019, a restaging PET/CT scan documented enlargement of the right middle lobe metastasis to 3.6 cm with increased hypermetabolism and a new right hilar node. He was started on pembrolizumab 200 mg IV every 2 weeks. In September 2019, he developed generalized arthralgia, which responded to prednisone 10 mg daily and minimal analgesics. He also developed grade 1 maculo-papular rash that responded to a low-potency steroid cream. In October 2019, PET/CT scan showed stability of the lung mass and hilar node, with decreased hypermetabolism.

In January 2020, he reported dry mouth, affecting his ability to chew and swallow food. He subsequently developed dental cavities requiring multiple dental visits for fillings and extractions. On physical examination, he had glossy appearing oral mucosa with geographical tongue and generalized shortened papillae in addition to multiple cavities, yielding a score of 9 on the Challacombe scale of clinical oral dryness. He also had mildly dry eyes that were well managed with artificial tears. The unstimulated whole salivary flow rate was less than 0.1 ml/min. The Schirmer test was positive (filter paper remained less than 5 mm of wetting after 5 min of continuous contact). ANA, RF, HIV, HCV, SS-A and SS-B antibodies were negative. Patient had an unremarkable bilateral parotid ultrasound and declined biopsy. He started using artificial saliva and lemon drops in addition to prednisone 10 mg daily, with no measurable improvement.

In May 2022, pembrolizumab dose was decreased to 100 mg every 2 weeks, which resulted in some improvement in his ability to chew and swallow. He developed a total of 17 dental cavities, required two extractions. The ongoing dental issues adversely affected his sense of well being and led to a significant financial burden. In July 2020, a PET/CT scan showed continued partial response to pembrolizumab. In October 2022, he was in complete remission but continued to have significant dry mouth leading to tooth loss. The decision was made to stop pembrolizumab in December 2022, after which his symptoms improved, and he did not develop any additional cavities. At a 6-month follow up, he remained in complete remission, but continued to suffer from clinically significant dry mouth requiring artificial saliva before meals. The calculated cause-effect probability using the Naranjo nomogram showed the relationship between pembrolizumab and the patient symptoms to be probable (Table 3).

OPEN IN VIEWER

Table 3.

The Naranjo Adverse Drug Reaction (ADR) probability scale questionnaire in the patient #3.

To assess the adverse drug reaction, please answer the following questionnaire and give the pertinent score.	Yes	No	Do not know	Score
1. Are there previous conclusive reports on this reaction?	+1	0	0	+1
2. Did the adverse event appear after the suspected drug was administered?	+2	−1	0	+2
3. Did the adverse event improve when the drug was discontinued or a specific antagonist was administered?	+1	0	0	+1
4. Did the adverse event reappear when the drug was readministered?	+2	−1	0	0
5. Are there alternative causes that could on their own have caused the reaction?	−1	+2	0	+2
6. Did the reaction reappear when a placebo was given?	−1	+1	0	0
7. Was the drug detected in the blood or other fluids in concentrations known to be toxic?	+1	0	0	0
8. Was the reaction more severe when the dose was increased or less severe when the dose was decreased?	+1	0	0	+1
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0
10. Was the adverse event confirmed by any objective evidence?	+1	0	0	+1
Score: 8

The Naranjo criteria classify the probability that an adverse event is related to drug therapy based on a list of weighted questions, which examine factors such as the temporal association of drug administration and event occurrence, alternative causes for the event, drug levels, dose-response relationships and previous patient experience with the medication. The ADR is assigned to a probability category from the total score as follows: definite if the overall score is 9 or greater, probable for a score of 5–8, possible for 1–4, and doubtful if the score is 0. The Naranjo criteria do not take into account drug-drug interactions. Drugs are evaluated individually for causality, and points are deducted if another factor may have resulted in the adverse event, thereby weakening the causal association.

Case 4

A 71-year-old Caucasian man was originally found to have a right lower lung nodule on low-dose CT scan for lung cancer screening. Biopsy showed adenocarcinoma believed to be primary in the lung. He had a right lower lobectomy and lymph node dissection in January 2023. The size of the primary tumor was 3.8 cm, and 1/15 excised nodes (station 7) were positive; consistent with TNM stage IIIA (pT2pN2M0) adenocarcinoma of the lung. Lympho-vascular invasion and visceral pleural involvement were detected; margins were negative. PD-L1 expression was 10%; no actionable mutations were detected. A PET/CT scan revealed no evidence of distant metastatic disease. He had four cycles of cisplatin-pemetrexed in from March through May 2023, followed by cross-maintenance with atezolizumab every 3 weeks, with a plan to complete 12 months. He had a personal history of chronic obstructive pulmonary disease, controlled hypertension, hyperlipidemia, coronary artery disease, peripheral artery disease and abdominal aortic aneurysm. His family history was unremarkable. Social history was significant for a 20-pack-year history of smoking, stopping in 2001; he did not consume alcohol in excess. In December 2023, a PET/CT scan showed no evidence of disease relapse. At approximately the same time, he reported the onset of mouth dryness. His symptoms progressed rapidly despite using artificial saliva and lemon drops for stimulating saliva production. Physical examination revealed slight conjunctival erythema and dry oral mucosa. There was no pooling of saliva in the floor of the mouth yielding a score of 4 on the Challacombe scale of clinical oral dryness. The Schirmer test was also positive (filter paper with less than 5 mm of wetting after 5 min of continuous contact). ANA, C3 and C4 complement, RF, SS-A and SS-B antibodies were all negative. He continued using artificial saliva. He continued atezolizumab 1200 mg IV every 3 weeks without interruption.

The calculated cause-effect probability using the Naranjo nomogram showed the relationship between atezolizumab and the patient symptoms to be probable (Table 4).

OPEN IN VIEWER

Table 4.

The Naranjo Adverse Drug Reaction (ADR) probability scale questionnaire in the patient #4.

To assess the adverse drug reaction, please answer the following questionnaire and give the pertinent score.	Yes	No	Do not know	Score
1. Are there previous conclusive reports on this reaction?	+1	0	0	+1
2. Did the adverse event appear after the suspected drug was administered?	+2	−1	0	+2
3. Did the adverse event improve when the drug was discontinued or a specific antagonist was administered?	+1	0	0	0
4. Did the adverse event reappear when the drug was readministered?	+2	−1	0	0
5. Are there alternative causes that could on their own have caused the reaction?	−1	+2	0	+2
6. Did the reaction reappear when a placebo was given?	−1	+1	0	0
7. Was the drug detected in the blood or other fluids in concentrations known to be toxic?	+1	0	0	0
8. Was the reaction more severe when the dose was increased or less severe when the dose was decreased?	+1	0	0	0
9. Did the patient have a similar reaction to the same or similar drugs in any previous exposure?	+1	0	0	0
10. Was the adverse event confirmed by any objective evidence?	+1	0	0	+1
Score: 6

The Naranjo criteria classify the probability that an adverse event is related to drug therapy based on a list of weighted questions, which examine factors such as the temporal association of drug administration and event occurrence, alternative causes for the event, drug levels, dose-response relationships and previous patient experience with the medication. The ADR is assigned to a probability category from the total score as follows: definite if the overall score is 9 or greater, probable for a score of 5–8, possible for 1–4, and doubtful if the score is 0. The Naranjo criteria do not take into account drug-drug interactions. Drugs are evaluated individually for causality, and points are deducted if another factor may have resulted in the adverse event, thereby weakening the causal association.