Treatment of induced oligometastatic disease after partial response to immunochemotherapy in patient with stage IV non-small cell lung cancer and severe toxicity

Abstract

Treatment of induced oligometastatic disease after partial response to systemic antineoplastic therapy in non-small cell lung cancer (NSCLC) remains controversial. The introduction of immune checkpoint inhibitors (ICIs) has revolutionised the treatment of stage IV NSCLC. While ICI combined with chemotherapy (ChT) leads to longer duration of response and higher response rates compared with ChT alone, it can also cause serious adverse events (AEs) resulting in treatment discontinuation. In case of treatment discontinuation due to AEs after partial response to systemic treatment, surgical treatment of residual disease can be considered as it could lead to complete response. We present a case of a patient with stage IV NSCLC who is currently alive without any signs of cancer after partial response to ICI/ChT followed by surgical removal of residual disease.

Background

Lung cancer is the leading cause of cancer death worldwide.¹ Non-small cell lung cancer (NSCLC) accounts for approximately 80% of all lung cancer cases, of which the majority are adenocarcinomas.² The prognosis of NSCLC depends primarily on the tumour, lymph nodes and metastasis (TNM) stage³ at the time of diagnosis, and 1 year survival rates vary from 59–88% for early stages to 16–26% for stage IV disease.⁴ Unfortunately, more than half of the patients are diagnosed at stage IV allowing only for palliative systemic therapy.⁴

The introduction of immune checkpoint inhibitors (ICIs) has revolutionised the treatment of stage IV NSCLC. Pembrolizumab, an anti-programmed death-1 (PD-1) monoclonal antibody, has proven to be effective in first-line treatment of advanced NSCLC as monotherapy in patients with a programmed death-ligand 1 (PD-L1) expression ≥50%⁵ and as triple therapy in combination with platinum-based chemotherapy (ChT) irrespective of the PD-L1 expression.^{6 7} In Denmark, pembrolizumab administered concomitant with pemetrexed and platinum-based ChT is an approved first-line palliative treatment for patients with performance status (PS) 0–1, non-squamous NSCLC, PD-L1 expression <50% and no targetable genetic aberrations. This triple therapy is administered every 3 weeks for 4 cycles followed by maintenance treatment with pembrolizumab and pemetrexed.⁸ The approval of the triple therapy was based on results from KEYNOTE-189, a randomised controlled trial (RCT), including previously untreated patients with stage IV non-squamous NSCLC. In this study, significantly improved objective response rate (ORR), median overall survival (OS) and progression free survival (PFS) were observed in patients receiving pembrolizumab combined with pemetrexed and platinum-based ChT (ICI/ChT) compared with those receiving placebo plus pemetrexed and platinum-based ChT (placebo/ChT). Additionally, the mean duration of response was longer among patients who received ICI/ChT, and more patients became long-term responders.⁹

Despite the greater response to ICI/ChT compared with placebo/ChT, the occurrence of immune-related adverse events (irAEs) was a reason for ICI discontinuation in 20.2% of patients.⁹ Treatment discontinuation due to adverse events often leads to a dilemma of whether or not to reinduce treatment in case of disease progression in the follow-up period. Similarly, treatment of residual disease of oligometastatic disease (OMD) can be challenging in patients whose treatment discontinuation was necessary due to irAEs, since ICI reinduction is associated with a significant risk of irAE recurrence.¹⁰ In such cases where ICI reinduction is considered too hazardous, surgical treatment could be a viable option.

We present a case of a patient with stage IV NSCLC who is currently alive without any signs of cancer 3 years after diagnosis due to partial response to ICI/ChT followed by surgical removal of residual disease.

Case presentation

A woman in her 50s with hypothyroidism and asthma, a former smoker with PS 0, presented with coughing and activity-related dyspnoea, which persisted despite treatment with different asthma medications. The patient also received antibiotics for pneumonia two times within a year without any improvement of the respiratory symptoms. Thus, the patient was referred to the Department of Respiratory Diseases by her general practitioner.

Investigations

The patient had normal spirometry. However, a CT scan revealed a large central tumour in the right lung, measuring 50 mm, which confluenced with lymph node metastases localised in station 11R (figure 1). Subsequently, the patient underwent a positron emission tomography–CT (PET/CT) scan and an endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA). The PET/CT demonstrated high ¹⁸F-fluorodeoxyglucose (FDG) uptake in cervical, periclavicular, mediastinal, retrocrural and periaortic lymph nodes as well as the lymph nodes by the right internal thoracic artery (figure 2). Thus, the patient was diagnosed with a stage IV adenocarcinoma of the right lung, T4N3M1c, with a PD-L1 expression of 40%. Epidermal growth factor receptor mutations, anaplastic lymphoma kinase translocation, ROS proto-oncogene 1 (ROS1) rearrangement or BRAF mutation were not observed.

Figure 1.

(A) Baseline CT scan, right hilar lesion (50 mm). (B) Radiologic response assessed by RECIST V.1.1¹² after 3 months of systemic treatment, right hilar lesion (39 mm, 22% reduction in target measurement). (C) Radiologic response assessed by RECIST V.1.1 1 month after last systemic treatment, right hilar lesion (28 mm, 44% reduction in target measurement). (D) Radiologic response assessed by RECIST V.1.1 3 months after last systemic treatment, right hilar lesion (27 mm, 46% reduction in target measurement). (E) Radiologic response assessed by RECIST V.1.1 8 months after last systemic treatment, right hilar lesion (24 mm, 52% reduction in target measurement). (F) Radiologic response assessed by RECIST V.1.1 2 years after last systemic treatment, right hilar lesion (non-measurable target). RECIST, Response Evaluation Criteria in Solid Tumours.

Figure 2.

(A) Baseline PET/CT, bilateral FDG-uptake in cervical, supraclavicular and mediastinal lymph nodes. (B) Baseline PET/CT, FDG-uptake in right hilar lesion. (C) Baseline PET/CT, FDG-uptake in retroperitoneal lymph node. (D) Metabolic response 12 months after last systemic treatment, FDG-uptake in left cervical lymph nodes. (E) Metabolic response 12 months after last systemic treatment, moderate FDG-uptake in right hilar lesion. (F) Metabolic response 12 months after last systemic treatment, retroperitoneal lymph nodes without FDG-uptake. FDG, ¹⁸F-fluorodeoxyglucose; PET/CT, positron emission tomography–CT.

Treatment

The patient was referred to the Department of Oncology, and palliative treatment with pembrolizumab in combination with pemetrexed and carboplatin was initiated. From the second cycle onwards, the doses of both pemetrexed and carboplatin were reduced to 75% of the recommended dose due to febrile neutropenia. After 7 cycles of treatment, the patient’s alanine aminotransferase increased to 83 U/L, corresponding to grade 1 hepatic impairment according to the Common Terminology Criteria for Adverse Events (CTCAE) V.5.0.¹¹ Shortly after, the patient began to experience severe CTCAE grade 3 peripheral sensory neuropathy, limiting self-care activities of daily living. As a result, the treatment was discontinued after a total of eight cycles. Steroid treatment was not initiated, as the symptoms gradually decreased 3–4 weeks after the last treatment cycle without medical intervention. However, the patient was recommended to discontinue the treatment and await disease response evaluation on an upcoming CT scan.

Outcome and follow-up

The treatment response was estimated according to the Response Evaluation Criteria in Solid Tumours (RECIST) V.1.1.¹² Compared with baseline PET/CT scan measurements, a primary tumour reduction of 22% after 3 treatment cycles was observed, consistent with stable disease. After the treatment was discontinued, the first follow-up CT scan showed 44% primary tumour reduction compared with the baseline measurements, consistent with partial response (PR). The patient had partially recovered from peripheral sensory neuropathy, but the treatment was still on hold. The patient reached full recovery from the neuropathy within 6 months of the last treatment. Subsequent CT scans, which were performed every 3 months, showed that the tumour reduction proceeded. Eight months after the last systemic treatment, the tumour was no longer measurable, but a residual lesion was observed corresponding to the localisation of the primary tumour, which confluenced with lymph node metastases in station 11R (figure 1). There were no signs of lymphadenopathy in the cervical, thoracic or abdominal regions.

Twelve months after the last systemic treatment, a PET/CT scan was conducted in order to assess the metabolic activity of the residual disease. The PET/CT scan showed high FDG-uptake in three cervical lymph nodes on the left side and moderate FDG-uptake in lymph nodes in stations 4R and 11R (figure 2). A fine-needle aspiration of one of the cervical lymph nodes showed findings consistent with a metastasis from the adenocarcinoma of the lung. An EBUS-TBNA of the mediastinal lymph nodes in stations 4R and 11R was inconclusive due to a low number of cells but showed no signs of malignancy. The EBUS was not repeated due to the normal lymph node morphology during the bronchoscopic ultrasound examination and the normal lymph node size on the PET/CT scan. Thus, it was concluded that the patient had a complete thoracic response with residual disease in the neck, consistent with induced OMD. The patient was proposed to consider active surveillance regarding the metastases in the cervical lymph nodes. However, reinduction of immunotherapy in case of progressive disease was not a preferable option for the patient. Therefore, surgery of the residual disease in the cervical lymph nodes was proposed and performed. The pathology report revealed that 4 of the 10 cervical lymph nodes, which were removed, contained metastases from the lung adenocarcinoma. No extranodal extension was observed. Therefore, postoperative radiation therapy was not considered indicated. Since the surgery, 24 months after the last systemic treatment, the patient has had no signs of clinical or radiological progression.

Discussion

Adenocarcinomas are the most prevalent histopathological type of NSCLC, and the incidence has increased over the last few decades.² Somatic alterations are present in approximately 12%–62% of cases and play an unquestionable role in development of cancer, especially in never-smokers.¹³ In Denmark and many other European countries, PD-L1 is the only approved biomarker for ICI and ICI/ChT in lung cancer treatment. However, intratumoral and intertumoral heterogeneity of PD-L1 expression is common and contributes to the weak predictive potential of PD-L1.^{14 15} This variability of PD-L1 expression within the tumour and between the tumour and distant metastases provides a possible explanation of the so-called mixed response observed across different tumour sites. The role of PD-L1 as a predictor of response to ICI is further called into question by the observation that PD-L1 expression can vary significantly between different assay methods.¹⁶ Thus, the predictive potential of PD-L1 expression is uncertain.

The results from KEYNOTE-189 demonstrated that ICI/ChT significantly improved ORR, OS and PFS compared with ChT alone irrespective of PD-L1 expression.⁹ Notably, 1.2% of the patients in the ICI/ChT group showed complete response. Moreover, a greater number of patients in the ICI/ChT group became long-term responders with estimated PFS rates at 24 months of 22.0% in the ICI/ChT group versus 3.4% in the placebo/ChT group.⁹

In patients treated with ICI/ChT, one of the greatest treatment-limiting factors is irAEs. In a real-world study, irAEs were the second most common reason for ICI discontinuation (18%) in patients who received first-line ICI/ChT.¹⁷ In KEYNOTE-189, the incidence of treatment-related AEs of grade ≥3 was higher in patients treated with ICI/ChT compared with patients who received placebo/ChT (52.3% vs 41.6%).⁹ Discontinuation of any treatment component due to treatment-related AEs was also more common in the ICI/ChT group (27.2% vs 9.4%). However, the median duration of treatment was 7.2 months in the ICI/ChT group and 4.2 months in the placebo/ChT group, and the exposure-adjusted AE rates were lower in patients who received ICI/ChT versus those who received placebo/ChT. This suggests that the increased toxicity observed in the ICI/ChT group was a result of increased treatment duration.⁹ Furthermore, a pooled analysis, including patients aged ≥18 years with advanced NSCLC with PD-L1-positive tumours from three RCTs, KEYNOTE-010, KEYNOTE-024 and KEYNOTE-042, showed that treatment with ICI compared with ChT led to fewer AEs of grade ≥3 (17.7% vs 41.2%), and treatment discontinuation was also less common (7.2% vs 9.3%).¹⁸ Thus, it can be speculated that treatment with ICI alone compared with ChT does not lead to more toxicity. In patients aged above 75 years, ICI had a more favourable safety profile than ChT.¹⁸

In the case of our patient, the combination of ICI and ChT led to hepatic and neurological toxicity. However, immunosuppressive treatment of the AEs was not necessary, as the symptoms spontaneously resolved after treatment discontinuation. Neurological irAEs of grade ≥3 are rarely observed (<1%), and strategies for optimal management of these irAEs are lacking in the literature.¹⁹ As observed in a Danish real-world study, the median time to treatment discontinuation (TTD) was 4.8 months in patients treated with first-line ICI monotherapy, corresponding to the TTD of our patient.²⁰

Induced OMD is characterised by a maximal five residual lesions in maximum three organs excluding mediastinum with a previous history of polymetastatic disease responding to systemic treatment.²¹ Treatment of residual lesions of induced oligometastatic NSCLC after PR to systemic antineoplastic therapy remains controversial. Whether to recommend active surveillance or salvage surgery depends on the reason for treatment discontinuation, risk of AE recurrence in case of treatment reinduction, salvage surgery possibilities and patient preferences. In our patient, the risk of reinducing severe peripheral sensory neuropathy after reinduction of ICI/ChT was the primary reason for surgical treatment of the residual disease. The biopsies of the FDG-uptake lesions were performed before surgery, and the patient was informed about the lack of randomised trials comparing active surveillance and salvage surgery in patients with AEs of grade ≥3 after ICI/ChT.

There are no prospective randomised trials investigating survival benefit of salvage surgery after PR to ICI or ICI/ChT. Surgical treatment of lung residual lesions after ICI has only been retrospectively analysed in a small cohort of 19 patients, wherein 7 of the 9 patients diagnosed with primary lung cancer had stage IV disease.²² Complete resection was achieved in 95% of the patients. The 2-year OS rate was 77%. However, the cohort also included patients with primary melanoma and three other solid tumour types.²² The role of surgical intervention in stage IV NSCLC patients remains disputable, and to the best our knowledge, only few case reports resembling treatment of our patient has been described to date.²³ Currently on-going phase III studies will provide an answer regarding local consolidation therapy in OMD compared with continuous systemic therapy (LONESTAR (NCT03391869), OMEGA (NCT03827577), OITROLC (NCT02076477) and SARON (NCT02417662)). However, those studies will not be directly applicable in our patient’s situation where toxicity was a reason for treatment discontinuation despite radiological response. Therefore, we find it necessary to share experience and gain evidence regarding salvage surgery after treatment with ICI or ICI/ChT.

Currently, the criteria for consideration of salvage surgery of residual disease or oligometastatic lesions are not clarified.²² However, in the era of possible long-term responses to ICI or ICI/ChT, surgery of residual disease in patients experiencing severe toxicity is expected to become more common in the future. According to an exploratory analysis of the pooled data from 12 RCTs, an ORR of 61% was observed in patients with NSCLC and PD-L1 ≥50% treated with first-line ICI/ChT compared with 43% in ICI alone (OR 1.2, 95% CI 1.1 to 1.3).²⁴ Thus, patients treated with ICI/ChT may have higher chances for greater local response, leading to better possibilities for surgical approach in case of treatment-limiting toxicity. This is of high importance, as toxicity is comparable in patients treated with ICI and ICI/ChT.⁹

As of now, 3 years after the diagnosis, our patient is still alive and has returned to her job. Treatment with ICI/ChT and active surveillance after development of irAEs resulted in partial radiological response, and the residual disease of the neck was successfully treated with surgery. The hilar lesion in the primary tumour site showed no signs of malignancy in the mediastinal rebiopsy and is radiologically stable though not measurable due to its small size. Based on this, it can be concluded that the patient is in complete remission, and the residual lesion most likely represents either necrosis, scar tissue or chronic inflammation (or a combination hereof) rather than residual hilar tumour.

We find this case important to share as it emphasises the possibility of salvage surgery in patients with stage IV NSCLC who show PR to systemic antineoplastic therapy and experience concurrent serious AEs leading to treatment discontinuation.

Patient’s perspective.

The following text has been translated from the original language (Danish) and approved by the patient.

On (date), I was diagnosed with incurable lung cancer. I asked the doctor about my life expectancy to which the doctor replied one to 2 years. I was then referred to the Department of Oncology for palliative treatment. My family and I were invited to a consultation at the Department of Oncology in early (date). We received a thorough explanation of the treatment course, and we felt that there was time and space for our questions and anxiety. After the consultation with the doctor, we talked to the nurse for an hour. I started treatment with immunotherapy and chemotherapy a few days later.

After the first treatment, I experienced extreme fatigue, hot flashes that resulted in soaked clothes in the neck and chest regions, palpitations, mild nausea, and mild constipation. On (date), I got a fever with temperatures varying between 38.0 °C and 39.7 °C. I contacted the Department of Oncology in (town) and was admitted at a local hospital, where I received antibiotic treatment. I was admitted for 4 days, and my next treatment was delayed. Because of the risks associated with the treatment, the doctor decided to reduce my chemotherapy to 75% of the original dose for all future cycles.

In the first 3–5 days after every treatment, I had mild nausea and constipation, which was treated with medicine. However, fatigue is the symptom that affected me the most during the treatment. As a result, simple household chores became difficult for me, for example, changing the bed sheets could take 2.5 hours because I needed many breaks. Social interactions also required a lot of energy and resulted in exhaustion. I had to tell my guests in advance that I prefer that the visits are not too long since I couldn’t predict how tired I would be after the visits and it was difficult for me to ask the guests to leave.

Towards the end of the treatment course, I began to experience numbness and pain in my hands. I found it difficult to hold things in my hands and found myself accidentally dropping cups of coffee among other things. It was also difficult for me to type messages on my iPad and iPhone, so I had to record voice messages instead. In (date), I had the metastases on my neck removed surgically. Since the surgery, I have experienced numbness in my shoulder and neck. I also had numbness in my face, which has disappeared now.

After the corona virus hit (country), my contact with the Department of Oncology changed. I assume that the regular consultations were replaced by phone consultations in order to protect both patients and doctors. However, the doctors often called the day before the scheduled phone consultation. Even though it was good news, these phone calls gave me anxiety because I was always home alone since my family had taken off on the day that the phone consultations were scheduled to take place. It was also difficult for me because I was called by different doctors who had to get to know me, and we would have to reach a mutual understanding. After about three phone consultations, I asked for a consultation face-to-face so that I could bring a family member. During this consultation, I expressed my anxiety and frustration about having to speak to a new doctor every time. It was therefore decided that my future consultations would be face-to-face, and I would see the same doctor every time if possible. This has given me a lot of peace and comfort.

I still experience fatigue if I push myself too hard at work and after excessive social interaction. This fatigue that I have experienced throughout the whole course has been a source of huge frustration. It is difficult for me to find a balance between wanting to work as I did before my diagnosis while also wanting to have enough energy left over for my family and friends. At the moment, I work between 15 hours and 21 hours a week. The cancer diagnosis has affected my psychological well-being a lot. During the first lockdown because of the COVID-19 pandemic, I felt anxiety and missed spending time with my children and grandchildren. I experienced forgetfulness and word-finding difficulties. When I decided to see my children and grandchildren despite the recommendations, I gradually began to feel better. After everything I have been through I have become more prone to crying, and even writing these last sentences brings tears in my eyes.

Learning points.

• Although adverse events of combination treatment with pembrolizumab plus pemetrexed and platinum-based chemotherapy can lead to treatment discontinuation, patients can still benefit from the treatment and show response several months after treatment discontinuation.

• Surgical treatment of residual disease in induced oligometastatic disease should be considered if possible in patients with partial response to pembrolizumab plus pemetrexed and platinum-based chemotherapy, as combined systemic and surgical treatment could lead to complete response.

• Prompt use of steroid treatment in adverse events to pembrolizumab plus pemetrexed and platinum-based chemotherapy should be avoided, especially in patients showing spontaneous recovery without steroid treatment.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s).