Durvalumab

The use of systemic corticosteroids or immunosuppressants before starting durvalumab, except physiological dose of systemic corticosteroids (≤10 mg/day prednisone or equivalent), is not recommended because of their potential interference with the pharmacodynamic activity and efficacy of durvalumab. However, systemic corticosteroids or other immunosuppressants can be used after starting durvalumab to treat immune-related adverse reactions.

Data from patients with severe renal impairment are too limited to draw conclusions on this population.

Data from patients with severe hepatic impairment are too limited to draw conclusions on this population.

In patients with pre-existing autoimmune disease (AID), data from observational studies suggest an increased risk of immune-related adverse reactions following immune-checkpoint inhibitor therapy as compared with patients without pre-existing AID. In addition, flares of the underlying AID were frequent, but the majority were mild and manageable.

There are no data on the use of durvalumab in pregnant women. Based on its mechanism of action, durvalumab has the potential to impact maintenance of pregnancy, and in a mouse allogeneic pregnancy model, disruption of PD-L1 signaling was shown to result in an increase in foetal loss. Animal studies with durvalumab are not indicative of reproductive toxicity. Human IgG1 is known to cross the placental barrier and placental transfer of durvalumab was confirmed in animal studies. Durvalumab may cause foetal harm when administered to a pregnant woman and is not recommended during pregnancy and in women of childbearing potential not using effective contraception during treatment and for at least 3 months after the last dose.

It is unknown whether durvalumab is secreted in human breast milk. Available toxicological data in cynomolgus monkeys have shown low levels of durvalumab in breast milk on day 28 after birth. In humans, antibodies may be transferred to breast milk, but the potential for absorption and harm to the newborn is unknown. However, a potential risk to the breast-fed child cannot be excluded. A decision must be made whether to discontinue breast-feeding or to discontinue or abstain from durvalumab therapy taking into account the benefit of breast-feeding for the child and the benefit of therapy for the woman.

Women of childbearing potential/Contraception

Women of childbearing potential should use effective contraception during treatment with durvalumab and for at least 3 months after the last dose of durvalumab.

Fertility

There are no data on the potential effects of durvalumab on fertility in humans or animals.

Durvalumab has no or negligible influence on the ability to drive and use machines.

Summary of the safety profile

Durvalumab as monotherapy

The safety of durvalumab as monotherapy is based on pooled data in 4 642 patients across multiple tumour types. Durvalumab was administered at a dose of 10 mg/kg every 2 weeks, 20 mg/kg every 4 weeks or 1 500 mg every 4 weeks. The most common (>10%) adverse reactions were cough/productive cough (18.1%), diarrhoea (15.1%), rash (15.0%), arthralgia (12.4%), pyrexia (12.5%), abdominal pain (11.8%), upper respiratory tract infections (11.8%), pruritus (11.1%), and hypothyroidism (11.6%). The most common (>2%) NCI CTCAE Grade ≥3 adverse reactions were pneumonia (3.4%) and aspartate aminotransferase increased/alanine aminotransferase increased (2.5%).

Durvalumab was discontinued due to adverse reactions in 3.9% of patients. The most common adverse reactions leading to treatment discontinuation were pneumonitis (1.1%) and pneumonia (0.8%).

Durvalumab was delayed or interrupted due to adverse reactions in 13.1% of patients. The most common adverse reactions leading to dose delay or interruption were pneumonia (2.3%) and aspartate aminotransferase increased/alanine aminotransferase increased (2.0%).

The safety of durvalumab as monotherapy in patients treated for HCC is based on data in 492 patients and was consistent with the overall safety profile in the durvalumab monotherapy pool (N=4 642). The most common (>10%) adverse reactions were AST increased/ALT increased (20.3%), abdominal pain (17.9%), diarrhoea (15.9%), pruritus (15.4%), and rash (15.2%). The most common (>2%) Grade ≥3 adverse reactions were AST increased/ALT increased (8.1%) and abdominal pain (2.2%).

Durvalumab was discontinued due to adverse reactions in 3.7% of patients. The most common adverse reactions leading to treatment discontinuation were AST increased/ALT increased (0.8%) and hepatitis (0.6%).

Durvalumab was delayed or interrupted due to adverse reactions in 11.6% of patients. The most common adverse reaction leading to dose delay or interruption was AST increased/ALT increased (5.9%).

Durvalumab in combination with chemotherapy

The safety of durvalumab in combination with chemotherapy is based on pooled data in 1 769 patients from 5 studies (TOPAZ-1, CASPIAN, DUO-E, AEGEAN and NIAGARA). The most common (>10%) adverse reactions were neutropenia (41.7%), anaemia (40.8%), nausea (40.1%), fatigue (39.6%), constipation (29.7%), decreased appetite (22.2%), thrombocytopenia (21.5%), alopecia (19.7%), rash (19.7%), diarrhoea (18.2%), vomiting (16.8%), abdominal pain (16.7%), neuropathy peripheral (16.3%), leukopenia (14.8%), pyrexia (14.0%), pruritus (13.0%), hypothyroidism (11.9%), arthralgia (11.5%), cough/productive cough (11.0%), and aspartate aminotransferase increased/alanine aminotransferase increased (10.7%). The most common (>2%) NCI CTCAE Grade ≥3 adverse reactions were neutropenia (25.2%), anaemia (13.7%), thrombocytopenia (6.9%), leukopenia (4.5%), fatigue (2.8%), pneumonia (2.4%) and febrile neutropenia (2.1%).

Durvalumab was discontinued due to adverse reactions in 6.2% of patients. The most common adverse reactions leading to treatment discontinuation were rash (0.7%), pneumonitis (0.7%) and fatigue (0.6%).

Durvalumab was delayed or interrupted due to adverse reactions in 29.2% of patients. The most common adverse reactions leading to dose delay or interruption were neutropenia (12.6%), thrombocytopenia (4.5%), anaemia (3.9%) and leukopenia (2.1%).

Durvalumab in combination with tremelimumab 75 mg and platinum-based chemotherapy

The safety of durvalumab given in combination with tremelimumab 75 mg and chemotherapy is based on data in 330 patients with metastatic NSCLC. The most common (>20%) adverse reactions were anaemia (49.7%), nausea (41.5%), neutropenia (41.2%), fatigue (36.1%), rash (25.8%), thrombocytopenia (24.5%) and diarrhoea (21.5%). The most common (>2%) NCI CTCAE Grade ≥3 adverse reactions were neutropenia (23.9%), anaemia (20.6%), pneumonia (9.4%), thrombocytopenia (8.2%), leukopenia (5.5%), fatigue (5.2%), lipase increased (3.9%), amylase increased (3.6%), febrile neutropenia (2.4%), colitis (2.1%) and aspartate aminotransferase increased/alanine aminotransferase increased (2.1%).

Durvalumab was discontinued due to adverse reactions in 8.5% of patients. The most common adverse reactions leading to treatment discontinuation were pneumonia (2.1%) and colitis (1.2%).

Durvalumab was interrupted due to adverse reactions in 49.4% of patients. The most common adverse reactions leading to dose interruption were neutropenia (16.1%), anaemia (10.3%), thrombocytopenia (7.3%), leukopenia (5.8%), pneumonia (5.2%), aspartate aminotransferase increased/alanine aminotransferase increased (4.8%), colitis (3.3%) and pneumonitis (3.3%).

Durvalumab in combination with tremelimumab 300 mg

The safety of durvalumab given in combination with a single dose of tremelimumab 300 mg is based on pooled data (HCC pool) in 462 HCC patients from the HIMALAYA Study and another study in HCC patients, Study 22. The most common (>10%) adverse reactions were rash (32.5%), pruritus (25.5%), diarrhoea (25.3%), abdominal pain (19.7%), aspartate aminotransferase increased/alanine aminotransferase increased (18.0%), pyrexia (13.9%), hypothyroidism (13.0%), cough/productive cough (10.8%), oedema peripheral (10.4%) and lipase increased (10.0%) (see Table 2). The most common severe adverse reactions (NCI CTCAE Grade ≥3) were aspartate aminotransferase increased/alanine aminotransferase increased (8.9%), lipase increased (7.1%), amylase increased (4.3%) and diarrhoea (3.9%).

The most common serious adverse reactions were colitis (2.6%), diarrhoea (2.4%), pneumonia (2.2%), and hepatitis (1.7%).

The frequency of treatment discontinuation due to adverse reactions was 6.5%. The most common adverse reactions leading to treatment discontinuation were hepatitis (1.5%) and aspartate aminotransferase increased/alanine aminotransferase increased (1.3%).

The severity of adverse drug reactions was assessed based on the CTCAE, defining grade 1=mild, grade 2=moderate, grade 3=severe, grade 4=life threatening and grade 5=death.

Durvalumab in combination with platinum-based chemotherapy followed by durvalumab in combination with olaparib 300 mg twice daily

The safety of durvalumab given in combination with platinum-based chemotherapy followed by durvalumab in combination with olaparib 300 mg twice daily is based on data in 238 patients with endometrial cancer. The most common (>20%) adverse reactions were anaemia (61.8%), nausea (54.6%), fatigue (54.2%), neuropathy peripheral (51.7%), alopecia (50.8%), neutropenia (39.5%), constipation (32.8%), thrombocytopenia (29.8%), diarrhoea (28.2%), vomiting (25.6%), arthralgia (24.4%), rash (23.5%), abdominal pain (23.5%), decreased appetite (23.1%) and leukopenia (20.2%).

The most common (>2%) NCI CTCAE Grade ≥3 adverse reactions were neutropenia (25.2%), anaemia (23.5%), leukopenia (6.7%), thrombocytopenia (5.9%), fatigue (5.5%), febrile neutropenia (3.4%), nausea (2.9%), aspartate aminotransferase increased/alanine aminotransferase increased (2.9%) and neuropathy peripheral (2.5%).

Durvalumab was discontinued in 4.6% of patients. The most common adverse reaction leading to treatment discontinuation was pneumonitis (1.7%).

Durvalumab was interrupted in 38.2% of patients. The most common adverse reactions leading to dose interruption were anaemia (13.4%), thrombocytopenia (11.8%), neutropenia (10.1%), leukopenia (2.9%), hypothyroidism (2.1%) and upper respiratory tract infection (2.1%).

Tabulated list of adverse reactions

Table 1 lists the incidence of adverse reactions in the durvalumab monotherapy pooled safety dataset (N=4 642), in patients treated with durvalumab in combination with chemotherapy (N=1 769) and in patients treated with durvalumab in combination with platinum-based chemotherapy followed by durvalumab in combination with olaparib (platinum-based chemotherapy + durvalumab + olaparib) (N=238). Unless otherwise stated, Table 2 lists the incidence of adverse reactions in patients treated with durvalumab in combination with tremelimumab 75 mg and platinum-based chemotherapy in the POSEIDON study (N=330) and in patients treated with durvalumab in combination with a single dose of tremelimumab 300 mg in the HCC pool (N=462). Adverse reactions are listed according to system organ class in MedDRA. Within each system organ class, the adverse reactions are presented in decreasing frequency. The corresponding frequency category for each ADR is defined as: very common (≥1/10); common (≥1/100 to <1/10); uncommon (≥1/1 000 to <1/100); rare (≥1/10 000 to <1/1 000); very rare (<1/10 000); not known (cannot be estimated from available data). Within each frequency grouping, adverse drug reactions are presented in order of decreasing seriousness.

Table 1. Adverse drug reactions in patients treated with durvalumab:

Adverse reaction frequencies may not be fully attributed to durvalumab alone but may contain contributions from the underlying disease or from other medicinal products used in a combination.
^* overall study of treatment with up to six 21-day cycles with platinum-based chemotherapy in combination with durvalumab, followed by durvalumab in combination with olaparib.
^a includes laryngitis, nasopharyngitis, peritonsillar abscess, pharyngitis, rhinitis, sinusitis, tonsillitis, tracheobronchitis and upper respiratory tract infection.
^b includes pneumocystis jirovecii pneumonia, pneumonia, pneumonia adenoviral, pneumonia bacterial, pneumonia cytomegaloviral, pneumonia haemophilus, pneumonia pneumococcal, pneumonia streptococcal, candida pneumonia, pneumonia klebsiella, and pneumonia legionella.
^c including fatal outcome.
^d includes gingivitis, oral infection, periodontitis, pulpitis dental, tooth abscess and tooth infection.
^e includes leukopenia and white blood cell count decreased.
^f includes neutropenia and neutrophil count decreased.
^g includes thrombocytopenia and platelet count decreased.
^h adverse reaction only applies to chemotherapy ADRs in the DUO-E study.
ⁱ adverse reaction only applies to olaparib ADRs in the DUO-E study.
^j includes drug hypersensitivity and hypersensitivity.
^k includes autoimmune hypothyroidism, hypothyroidism, immune-mediated hypothyroidism, blood thyroid stimulating hormone increased.
^l includes hyperthyroidism, Grave's disease, immune-mediated hyperthyroidism and blood thyroid stimulating hormone decreased.
^m includes autoimmune thyroiditis, immune-mediated thyroiditis, thyroiditis, and thyroiditis subacute.
ⁿ includes neuropathy peripheral, paraesthesia and peripheral sensory neuropathy.
° includes dysgeusia and taste disorder.
^p includes encephalitis, encephalitis autoimmune, immune-mediated encephalitis and noninfective encephalitis.
^q events were reported from post-marketing data.
^r includes deep vein thrombosis, embolism, embolism venous, pelvic venous thrombosis, superficial vein thrombosis and thrombosis.
^s includes dyspnoea and dyspnoea exertional.
^t includes pneumonitis and immune-mediated lung disease.
^u includes abdominal pain, abdominal pain lower, abdominal pain upper and flank pain.
^v includes stomatitis and mucosal inflammation.
^w includes colitis, enteritis, enterocolitis, immune-mediated enterocolitis and proctitis.
^x includes pancreatitis, pancreatitis acute, and immune-mediated pancreatitis.
^y includes alanine aminotransferase increased, aspartate aminotransferase in^creased, hepatic enzyme increased and transaminases increased.
^z includes hepatitis, autoimmune hepatitis, hepatitis toxic, hepatitis acute, hepatotoxicity, immune-mediated hepatitis, and hepatic cytolysis.
^aa includes rash erythematous, rash macular, rash maculopapular, rash papular, rash pruritic, rash pustular, erythema, eczema and rash.
^bb includes dermatitis and immune-mediated dermatitis.
^cc includes pemphigoid, dermatitis bullous and pemphigus. Reported frequency from completed and ongoing studies is uncommon.
^dd includes myositis and rhabdomyolysis.
^ee includes autoimmune arthritis, immune-mediate arthritis, polyarthritis, and rheumatoid arthritis.
^ff polymyositis (fatal) was observed in a patient treated with durvalumab from an ongoing sponsored clinical study outside of the pooled dataset.
^gg not observed in the durvalumab+Chemotherapy pool or the platinum-based chemotherapy+durvalumab+olaparib dataset, but observed in other AstraZeneca-sponsored clinical studies.
^hh includes autoimmune nephritis, tubulointerstitial nephritis, nephritis, glomerulonephritis, glomerulonephritis membranous, and immune-mediated nephritis.
ⁱⁱ includes fatigue and asthenia.
^jj includes oedema peripheral and peripheral swelling.
^kk includes infusion-related reaction and urticaria with onset on the day of dosing or 1 day after dosing.

Table 2. Adverse drug reactions in patients treated with durvalumab in combination with tremelimumab:

^a Includes laryngitis, nasopharyngitis, pharyngitis, rhinitis, sinusitis, tonsillitis, tracheobronchitis and upper respiratory tract infection.
^b Includes pneumocystis jirovecii pneumonia, pneumonia and pneumonia bacterial.
^c Includes periodontitis, pulpitis dental, tooth abscess and tooth infection.
^d Adverse reaction only applies to chemotherapy ADRs in the POSEIDON study.
^e Includes neutropenia and neutrophil count decreased.
^f Includes platelet count decreased and thrombocytopenia.
^g Includes leukopenia and white blood cell count decreased.
^h Adverse reaction was not observed in the HCC pool, but was reported in patients treated with durvalumab or durvalumab+tremelimumab in AstraZeneca-sponsored clinical studies.
ⁱ Includes blood thyroid stimulating hormone increased, hypothyroidism and immune-mediated hypothyroidism.
^j Includes blood thyroid stimulating hormone decreased and hyperthyroidism.
^k Includes autoimmune thyroiditis, immune-mediated thyroiditis, thyroiditis and thyroiditis subacute.
^l Includes neuropathy peripheral, paraesthesia and peripheral sensory neuropathy.
^m Includes encephalitis and encephalitis autoimmune.
ⁿ Adverse reaction was not observed in the POSEIDON study but was reported in patients treated with durvalumab or durvalumab+tremelimumab in clinical studies outside of the POSEIDON dataset.
° Reported in studies outside of the POSEIDON study and HCC pool.
^p Includes autoimmune myocarditis.
^q Includes immune-mediated pneumonitis and pneumonitis.
^r Includes abdominal pain, abdominal pain lower, abdominal pain upper and flank pain.
^s Includes mucosal inflammation and stomatitis.
^t Includes colitis, enteritis and enterocolitis.
^u Includes autoimmune pancreatitis, pancreatitis and pancreatitis acute.
^v Includes alanine aminotransferase increased, aspartate aminotransferase increased, hepatic enzyme increased and transaminases increased.
^w Includes autoimmune hepatitis, hepatitis, hepatocellular injury, hepatotoxicity, hepatitis acute and immune-mediated hepatitis.
^x Includes eczema, erythema, rash, rash macular, rash maculopapular, rash papular, rash pruritic and rash pustular.
^y Includes dermatitis and immune-mediated dermatitis.
^z Includes rhabdomyolysis, myositis, and polymyositis.
^aa Includes autoimmune nephritis and immune-mediated nephritis.
^bb Includes oedema peripheral and peripheral swelling.
^cc Includes infusion-related reaction and urticaria.

Description of selected adverse reactions

Durvalumab is associated with immune-mediated adverse reactions. Most of these, including severe reactions, resolved following initiation of appropriate medical therapy and/or treatment modifications.

The data for the following immune-mediated adverse reactions reflect the durvalumab monotherapy combined safety database of 4 642 patients which includes the PACIFIC, HIMALAYA and ADRIATIC studies and additional studies in patients with various solid tumours, in indications for which durvalumab is not approved. Across all studies, durvalumab was administered at a dose of 10 mg/kg every 2 weeks, 20 mg/kg every 4 weeks or 1 500 mg every 3 or 4 weeks. Details for the significant adverse reactions for durvalumab when given in combination with chemotherapy are presented if clinically relevant differences were noted in comparison to durvalumab monotherapy.

The data for the following immune-mediated adverse reactions are also based on 2 280 patients who received durvalumab 20 mg/kg every 4 weeks in combination with tremelimumab 1 mg/kg or durvalumab 1 500 mg in combination with tremelimumab 75 mg every 4 weeks. Details for the significant adverse reactions for durvalumab when given in combination with tremelimumab and platinum-based chemotherapy are presented if clinically relevant differences were noted in comparison to durvalumab in combination with tremelimumab.

The data for the following immune-mediated adverse reactions also reflect the durvalumab in combination with tremelimumab 300 mg combined safety database of 462 patients with HCC (the HCC pool). In these two studies, durvalumab was administered at a dose of 1 500 mg in combination with tremelimumab 300 mg every 4 weeks.

Immune-mediated pneumonitis

In the combined safety database with durvalumab monotherapy, (n=4 642 multiple tumour types), immune-mediated pneumonitis occurred in 147 (3.2%) patients, including Grade 3 in 37 (0.8%) patients, Grade 4 in 2 (<0.1%) patients and Grade 5 in 10 (0.2%) patients. The median time to onset was 56 days (range: 1-1 308 days). One hundred and fourteen of the 147 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day) and 4 patients also received other immunosuppressants including infliximab and cyclosporine. durvalumab was discontinued in 60 patients. Resolution occurred in 85 patients.

Immune-mediated pneumonitis occurred more frequently in patients in the PACIFIC Study who had completed treatment with concurrent chemoradiation within 1 to 42 days prior to initiation of study treatment (10.7%), than in the other patients in the combined safety database (1.0%).

In the PACIFIC Study, (n=475 in the durvalumab arm, and n=234 in the placebo arm) immune-mediated pneumonitis occurred in 47 (9.9%) patients in the durvalumab-treated group and 14 (6.0%) patients in the placebo group, including Grade 3 in 9 (1.9%) patients on durvalumab vs. 6 (2.6%) patients on placebo and Grade 5 (fatal) in 4 (0.8%) patients on durvalumab vs. 3 (1.3%) patients on placebo. The median time to onset in the durvalumab-treated group was 46 days (range: 2-342 days) vs. 57 days (range: 26-253 days) in the placebo group. In the durvalumab-treated group, all patients received systemic corticosteroids, including 30 patients who received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day) and 2 patients also received infliximab. In the placebo group, all patients received systemic corticosteroids, including 12 patients who received high-dose corticosteroid treatment and 1 patient also received cyclophosphamide and tacrolimus. Resolution occurred for 29 patients in the durvalumab treated group vs. 6 in placebo.

In the ADRIATIC Study, in patients with LS-SCLC (n=262 in the durvalumab arm, and n=265 in the placebo arm), immune-mediated pneumonitis occurred in 31 (11.8%) patients in the durvalumab-treated group and 8 (3.0%) patients in the placebo group, including Grade 3 in 5 (1.9%) patients on durvalumab vs. 1 (0.4%) patient on placebo and Grade 5 (fatal) in 1 (0.4%) patient on durvalumab. The median time to onset in the durvalumab-treated group was 55 days (range: 1-375 days) vs. 65.5 days (range: 24-124 days) in the placebo group. In the durvalumab-treated group all patients received systemic corticosteroids, including 25 patients who received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day) and 1 patient also received infliximab. In the placebo group all patients received systemic corticosteroids, including 7 patients who received high-dose corticosteroid treatment. Resolution occurred for 18 patients in the durvalumab treated group vs. 3 in placebo.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated pneumonitis occurred in 86 (3.8%) patients, including Grade 3 in 30 (1.3%) patients, Grade 4 in 1 (<0.1%) patient, and Grade 5 (fatal) in 7 (0.3%) patients. The median time to onset was 57 days (range: 8 - 912 days). All patients received systemic corticosteroids and 79 of the 86 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Seven patients also received other immunosuppressants. Treatment was discontinued in 39 patients. Resolution occurred in 51 patients.

In the HCC pool (n=462), immune-mediated pneumonitis occurred in 6 (1.3%) patients, including Grade 3 in 1 (0.2%) patient and Grade 5 (fatal) in 1 (0.2%) patient. The median time to onset was 29 days (range: 5-774 days). Six patients received systemic corticosteroids, and 5 of the 6 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). One patient also received other immunosuppressants. Treatment was discontinued in 2 patients. Resolution occurred in 3 patients.

In the DUO-E Study, out of 238 patients treated with platinum-based chemotherapy in combination with durvalumab, followed by durvalumab in combination with olaparib (platinum-based chemotherapy + durvalumab + olaparib arm) immune-mediated pneumonitis occurred in 5 2.1%) patients, including Grade 3 in 3 (1.3%) patients. The median time to onset was 85 days (range: 65-321 days). Five patients received systemic corticosteroids, including 4 patients who received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Resolution occurred in all 5 patients.

Immune-mediated hepatitis

In the combined safety database with durvalumab monotherapy, immune-mediated hepatitis occurred in 120 (2.6%) patients, including Grade 3 in 70 (1.5%) patients, Grade 4 in 9 (0.2%) patients and Grade 5 (fatal) in 6 (0.1%) patients. The median time to onset was 36 days (range: 1-644 days). Ninety-four of the 120 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Nine patients also received other immunosuppressants including mycophenolate treatment. durvalumab was discontinued in 30 patients. Resolution occurred in 56 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated hepatitis occurred in 80 (3.5%) patients, including Grade 3 in 48 (2.1%) patients, Grade 4 in 8 (0.4%) patients and Grade 5 (fatal) in 2 (<0.1%) patients. The median time to onset was 36 days (range: 1 - 533 days). All patients received systemic corticosteroids and 68 of the 80 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Eight patients also received other immunosuppressants. Treatment was discontinued in 27 patients. Resolution occurred in 47 patients.

In the HCC pool (n=462), immune-mediated hepatitis occurred in 34 (7.4%) patients, including Grade 3 in 20 (4.3%) patients, Grade 4 in 1 (0.2%) patient and Grade 5 (fatal) in 3 (0.6%) patients. The median time to onset was 29 days (range: 13-313 days). All patients received systemic corticosteroids, and 32 of the 34 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Nine patients also received other immunosuppressants. Treatment was discontinued in 10 patients. Resolution occurred in 13 patients.

Immune-mediated colitis

In the combined safety database with durvalumab monotherapy, immune-mediated colitis or diarrhoea occurred in 79 (1.7%) patients, including Grade 3 in 15 (0.3%) patients and Grade 4 in 2 (<0.1%) patients. The median time to onset was 72 days (range: 1-920 days). Fifty-five of the 79 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Five patients also received other immunosuppressants including infliximab treatment and mycophenolate. durvalumab was discontinued in 15 patients. Resolution occurred in 54 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated colitis or diarrhoea occurred in 167 (7.3%) patients, including Grade 3 in 76 (3.3%) patients and Grade 4 in 3 (0.1%) patients. The median time to onset was 57 days range: 3-906 days). All patients received systemic corticosteroids and 151 of the 167 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Twenty-two patients also received other immunosuppressants. Treatment was discontinued in 54 patients. Resolution occurred in 141 patients.

Intestinal perforation and large intestine perforation were uncommonly reported in patients receiving durvalumab in combination with tremelimumab.

In the HCC pool (n=462), immune-mediated colitis or diarrhoea occurred in 31 (6.7%) patients, including Grade 3 in 17 (3.7%) patients. The median time to onset was 23 days (range: 2-479 days). All patients received systemic corticosteroids, and 28 of the 31 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Four patients also received other immunosuppressants. Treatment was discontinued in 5 patients. Resolution occurred in 29 patients.

Intestinal perforation was observed in patients receiving durvalumab in combination with tremelimumab (rare) in studies outside of the HCC pool.

Immune-mediated endocrinopathies

Immune-mediated hypothyroidism

In the combined safety database with durvalumab monotherapy, immune-mediated hypothyroidism occurred in 384 (8.3%) patients, including Grade 3 in 7 (0.2%) patients. The median time to onset was 90.5 days (range: 1-951 days). Of the 384 patients, 379 patients received hormone replacement therapy and 7 patients received high-dose corticosteroids (at least 40 mg prednisone or equivalent per day) for immune-mediated hypothyroidism. One patient discontinued durvalumab due to immune-mediated hypothyroidism. Resolution occurred in 79 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated hypothyroidism occurred in 209 (9.2%) patients, including Grade 3 in 6 (0.3%) patients. The median time to onset was 85 days (range: 1-624 days). Thirteen patients received systemic corticosteroids and 8 of the 13 received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Treatment discontinued in 3 patients. Resolution occurred in 52 patients. Immune-mediated hypothyroidism was preceded by immune-mediated hyperthyroidism in 25 patients or immune-mediated thyroiditis in 2 patients.

In the HCC pool (n=462), immune-mediated hypothyroidism occurred in 46 (10.0%) patients. The median time to onset was 85 days (range: 26-763 days). One patient received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). All patients required other therapy including hormone replacement therapy. Resolution occurred in 6 patients. Immune-mediated hypothyroidism was preceded by immune-mediated hyperthyroidism in 4 patients.

Immune-mediated hyperthyroidism

In the combined safety database with durvalumab monotherapy, immune-mediated hyperthyroidism occurred in 76 (1.6%) patients. The median time to onset was 43 days (range: 1-253 days). Seventy-one of the 76 patients received medical therapy (thiamazole, carbimazole, propylthiouracil, perchlorate, calcium channel blocker or beta-blocker), 15 patients received systemic corticosteroids and 8 of the 15 patients received high-dose systemic corticosteroid treatment (at least 40 mg prednisone or equivalent per day). One patient discontinued durvalumab due to immune-mediated hyperthyroidism. Resolution occurred in 62 patients. Thirty-one patients experienced hypothyroidism following hyperthyroidism.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated hyperthyroidism occurred in 62 (2.7%) patients, including Grade 3 in 5 (0.2%) patients. The median time to onset was 33 days (range: 4-176 days). Eighteen patients received systemic corticosteroids, and 11 of the 18 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Fifty-three patients required other therapy (thiamazole, carbimazole, propylthiouracil, perchlorate, calcium channel blocker or beta-blocker), One patient discontinued treatment due to hyperthyroidism. Resolution occurred in 47 patients.

In the HCC pool (n=462), immune-mediated hyperthyroidism occurred in 21 (4.5%) patients, including Grade 3 in 1 (0.2%) patient. The median time to onset was 30 days (range: 13-60 days). Four patients received systemic corticosteroids, and all of the four patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Twenty patients required other therapy (thiamazole, carbimazole, propylthiouracil, perchlorate, calcium channel blocker, or beta-blocker). One patient discontinued treatment due to hyperthyroidism. Resolution occurred in 17 patients.

Immune-mediated thyroiditis

In the combined safety database with durvalumab monotherapy, immune-mediated thyroiditis occurred in 21 (0.5%) patients, including Grade 3 in 2 (<0.1%) patients. The median time to onset was 57 days (range: 14-217 days). Of the 21 patients, 18 patients received hormone replacement therapy and 3 patients received high-dose corticosteroids (at least 40 mg prednisone or equivalent per day). One patient discontinued durvalumab due to immune-mediated thyroiditis. Resolution occurred in 8 patients. Five patients experienced hypothyroidism following thyroiditis.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated thyroiditis occurred in 15 (0.7%) patients, including Grade 3 in 1 (<0.1%) patient. The median time to onset was 57 days (range: 22-141 days). Five patients received systemic corticosteroids and 2 of the 5 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Thirteen patients required other therapy including, hormone replacement therapy, thiamazole, carbimazole, propylthiouracil, perchlorate, calcium channel blocker, or beta-blocker. No patients discontinued treatment due to immune-mediated thyroiditis. Resolution occurred in 5 patients.

In the HCC pool (n=462), immune-mediated thyroiditis occurred in 6 (1.3%) patients. The median time to onset was 56 days (range: 7-84 days). Two patients received systemic corticosteroids, and 1 of the 2 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). All patients required other therapy including hormone replacement therapy. Resolution occurred in 2 patients.

Immune-mediated adrenal insufficiency

In the combined safety database with durvalumab monotherapy, immune-mediated adrenal insufficiency occurred in 24 (0.5%) patients, including Grade 3 in 8 (0.2%) patients. The median time to onset was 157.5 days (range: 20-547 days). All 24 patients received systemic corticosteroids; 8 of the 24 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). One patient discontinued durvalumab due to immune-mediated adrenal insufficiency. Resolution occurred in 6 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated adrenal insufficiency occurred in 33 (1.4%) patients, including Grade 3 in 16 (0.7%) patients and Grade 4 in 1 (<0.1%) patient. The median time to onset was 105 days (range: 20-428 days). Thirty-two patients received systemic corticosteroids, and 10 of the 32 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Treatment was discontinued in one patient. Resolution occurred in 11 patients.

In the HCC pool (n=462), immune-mediated adrenal insufficiency occurred in 6 (1.3%) patients, including Grade 3 in 1 (0.2%) patient. The median time to onset was 64 days (range: 43-504 days). All patients received systemic corticosteroids, and 1 of the 6 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Resolution occurred in 2 patients.

Immune-mediated type 1 diabetes mellitus

In the combined safety database with durvalumab monotherapy, immune-mediated type 1 diabetes mellitus occurred in 5 (0.1%) patients, including Grade 3 in 3 (0.1%) patients and Grade 4 in 1 (<0.1%) patient. The time to onset was 43 days (range: 29-631 days). All five patients required insulin therapy. durvalumab was permanently discontinued in one patient. One patient recovered and one patient recovered with sequelae.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated type 1 diabetes mellitus occurred in 6 (0.3%) patients, including Grade 3 in 1 (<0.1%) patient and Grade 4 in 2 (<0.1%) patients. The median time to onset was 58 days (range: 7-220 days). All patients required insulin. Treatment was discontinued for 1 patient. Resolution occurred in 1 patient.

Immune mediated hypophysitis/hypopituitarism

In the combined safety database with durvalumab monotherapy, immune-mediated hypophysitis/hypopituitarism occurred in 6 (0.1%) patients, including Grade 3 in 5 (0.1%) patients. The time to onset for the events was 85 days (range: 44-225 days). Three patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day), three patients discontinued durvalumab due to immune-mediated hypophysitis/hypopituitarism and resolution occurred in 1 patient.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated hypophysitis/hypopituitarism occurred in 16 (0.7%) patients, including Grade 3 in 8 (0.4%) patients. The median time to onset for the events was 123 days (range: 63-388 days). All patients received systemic corticosteroids and 8 of the 16 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Four patients also required endocrine therapy. Treatment was discontinued in 2 patients. Resolution occurred in 7 patients.

In the HCC pool (n=462), immune-mediated hypophysitis/hypopituitarism occurred in 5 (1.1%) patients. The median time to onset for the events was 149 days (range: 27-242 days). Four patients received systemic corticosteroids, and 1 of the 4 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Three patients also required endocrine therapy. Resolution occurred in 2 patients.

Immune-mediated nephritis

In the combined safety database with durvalumab monotherapy, immune-mediated nephritis occurred in 17 (0.4%) patients, including Grade 3 in 4 (0.1%) patients and Grade 4 in 1 (<0.1%) patient. The median time to onset was 84 days (range: 4-393 days). Twelve patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day) and 1 patient also received mycophenolate. durvalumab was discontinued in 7 patients. Resolution occurred in 8 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated nephritis occurred in 9 (0.4%) patients, including Grade 3 in 1 (<0.1%) patient. The median time to onset was 79 days (range: 39-183 days). All patients received systemic corticosteroids and 7 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Treatment was discontinued in 3 patients. Resolution occurred in 5 patients.

In the HCC pool (n=462), immune-mediated nephritis occurred in 4 (0.9%) patients, including Grade 3 in 2 (0.4%) patients. The median time to onset was 53 days (range: 26-242 days). All patients received systemic corticosteroids, and 3 of the 4 received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Treatment was discontinued in 2 patients. Resolution occurred in 3 patients.

Immune-mediated rash

In the combined safety database with durvalumab monotherapy, immune-mediated rash or dermatitis (including pemphigoid) occurred in 74 (1.6%) patients, including Grade 3 in 20 (0.4%) patients. The median time to onset was 56 days (range: 4-600 days). Thirty-seven of the 74 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). durvalumab was discontinued in 5 patients. Resolution occurred in 46 patients.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), immune-mediated rash or dermatitis (including pemphigoid) occurred in 112 (4.9%) patients, including Grade 3 in 17 (0.7%) patients. The median time to onset was 35 days (range: 1-778 days). All patients received systemic corticosteroids, and 57 of the 112 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Treatment was discontinued in 10 patients. Resolution occurred in 65 patients.

In the HCC pool (n=462), immune-mediated rash or dermatitis (including pemphigoid) occurred in 26 (5.6%) patients, including Grade 3 in 9 (1.9%) patients and Grade 4 in 1 (0.2%) patient. The median time to onset was 25 days (range: 2-933 days). All patients received systemic corticosteroids and 14 of the 26 patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). One patient received other immunosuppressants. Treatment was discontinued in 3 patients. Resolution occurred in 19 patients.

In the DUO-E Study, out of 238 patients treated with platinum-based chemotherapy in combination with durvalumab, followed by durvalumab in combination with olaparib (platinum-based chemotherapy + durvalumab + olaparib arm) immune-mediated rash occurred in 8 (3.4%) patients, including Grade 3 in 2 (0.8%) patients. The median time to onset was 155 days (range: 2-308 days). All patients received high-dose corticosteroid treatment (at least 40 mg prednisone or equivalent per day). Resolution occurred in all 8 patients.

Infusion-related reactions

In the combined safety database with durvalumab monotherapy, infusion-related reactions occurred in 70 (1.5%) patients, including Grade 3 in 6 (0.1%) patients. There were no Grade 4 or 5 events.

In the combined safety database with durvalumab in combination with tremelimumab (n=2 280), infusion-related reactions occurred in 45 (2.0%) patients, including Grade 3 in 2 (<0.1%) patients. There were no Grade 4 or 5 events.

In the DUO-E Study, out of 238 patients treated with platinum-based chemotherapy in combination with durvalumab, followed by durvalumab in combination with olaparib (platinum-based chemotherapy + durvalumab + olaparib arm), infusion-related reactions occurred in 13 (5.5%) patients, including Grade 3 in 1 (0.4%) patient. There were no Grade 4 or 5 events.

Pure Red Cell Aplasia

Pure Red Cell Aplasia (PRCA) has been reported when durvalumab has been used in combination with olaparib. In a clinical study of patients with endometrial cancer treated with durvalumab in combination with olaparib, the incidence of PRCA was 1.6%. All events were CTCAE Grade 3 or 4. Events were manageable following discontinuation of both durvalumab and olaparib. The majority of events were managed with blood transfusion and immunosuppression and recovered; there were no fatal events.

Laboratory abnormalities

In patients treated with durvalumab monotherapy, the proportion of patients who experienced a shift from baseline to a Grade 3 or 4 laboratory abnormality was as follows: 3.7% for alanine aminotransferase increased, 5.7% for aspartate aminotransferase increased, 0.9% for blood creatinine increased, 4.8% for amylase increased and 8.2% for lipase increased. The proportion of patients who experienced a TSH shift from baseline that was ≤ ULN to any grade > ULN was 20% and a TSH shift from baseline that was ≥ LLN to any grade < LLN was 18.2%.

In patients treated with durvalumab in combination with chemotherapy, the proportion of patients who experienced a shift from baseline to a Grade 3 or 4 laboratory abnormality was as follows: 4.6% for alanine aminotransferase increased, 3.9% for aspartate aminotransferase increased, 4.6% for blood creatinine increased, 5.7% for amylase increased, 10.2% for lipase increased, and 3.0% for bilirubin increased. The proportion of patients who experienced a TSH shift from baseline that was ≤ ULN to any grade > ULN was 23.1% and a TSH shift from baseline that was ≥ LLN to any grade < LLN was 21.6%.

In patients treated with durvalumab in combination with tremelimumab and platinum-based chemotherapy, the proportion of patients who experienced a shift from baseline to a Grade 3 or 4 laboratory abnormality was as follows: 6.2% for alanine aminotransferase increased, 5.2% for aspartate aminotransferase increased, 4.0% for blood creatinine increased, 9.4% for amylase increased and 13.6% for lipase increased. The proportion of patients who experienced a TSH shift from baseline that was ≤ ULN to > ULN was 24.8% and a TSH shift from baseline that was ≥ LLN to < LLN was 32.9%.

In patients treated with durvalumab in combination with tremelimumab, the proportion of patients who experienced a shift from baseline to a Grade 3 or 4 laboratory abnormality was as follows: 5.1% for alanine aminotransferase increased, 5.8% for aspartate aminotransferase, 1.0% for blood creatinine increased, 5.9% for amylase increased and 11.3% for lipase increased. The proportion of patients who experienced a TSH shift from baseline that was ≤ ULN to > ULN was 4.2% and a TSH shift from baseline that was ≥ LLN to < LLN was 17.2%.

In patients treated with platinum-based chemotherapy in combination with durvalumab, followed by durvalumab in combination with olaparib (platinum-based chemotherapy + durvalumab + olaparib arm), the proportion of patients who experienced a shift from baseline to a Grade 3 or 4 laboratory abnormality was as follows: 3.8% for alanine aminotransferase increased, 3.4% for aspartate aminotransferase increased and 1.7% for blood creatinine increased. The proportion of patients who experienced a TSH shift from baseline that was ≤ ULN to > ULN was 28.6% and a TSH shift from baseline that was ≥ LLN to < LLN was 20.1%.

Immunogenicity

Immunogenicity of durvalumab as monotherapy is based on pooled data in 3 069 patients who were treated with durvalumab 10 mg/kg every 2 weeks, or 20 mg/kg every 4 weeks as a single-agent and evaluable for the presence of anti-drug antibodies (ADAs). Eighty-four patients (2.7%) tested positive for treatment emergent ADAs. Neutralising antibodies (nAb) against durvalumab were detected in 0.5% (16/3 069) of patients. The presence of ADAs did not have a clinically relevant effect on pharmacokinetics or safety. There are insufficient number of patients to determine ADA impact on efficacy.

Across multiple phase III studies, in patients treated with durvalumab in combination with other therapeutic agents, 0% to 10.1% of patients developed treatment-emergent ADAs. Neutralizing antibodies against durvalumab were detected in 0% to 1.7% of patients treated with durvalumab in combination with other therapeutic agents. The presence of ADAs did not have an apparent effect on pharmacokinetics or safety.

Elderly

No overall differences in safety were reported between elderly (≥65 years) and younger patients.

In studies PACIFIC, ADRIATIC, CASPIAN, TOPAZ-1, HIMALAYA and NIAGARA data on safety for patients 75 years and older are too limited to draw a conclusion on this population.

In first line metastatic NSCLC patients in the POSEIDON study, some differences in safety were reported between elderly (≥65 years) and younger patients. The safety data from patients 75 years of age or older are limited to a total of 74 patients. There was a higher frequency of serious adverse reactions and discontinuation rate of any study treatment due to adverse reactions in 35 patients aged 75 years of age or older treated with durvalumab in combination with tremelimumab and platinum-based chemotherapy (45.7% and 28.6%, respectively) relative to 39 patients aged 75 years of age or older who received platinum-based chemotherapy only (35.9% and 20.5%, respectively).

In resectable NSCLC patients in the AEGEAN study, some differences in safety were reported between elderly (≥65 years) and younger patients. The safety data from patients 75 years of age or older are limited to 86 patients in both treatment arms. There was a higher frequency of serious adverse reactions in patients aged 75 years or older who received durvalumab in combination with chemotherapy relative to patients who received chemotherapy only (26.5% vs. 10.8%, respectively). There was a higher frequency of discontinuation of any study treatment due to adverse reactions in patients aged 75 years or older who received durvalumab in combination with chemotherapy relative to patients who received chemotherapy only (16.3% vs. 8.1%, respectively).