“Hockey Stick” may Strike Back: Hepatocellular Carcinoma on Noncirrhotic Liver as a Late Toxicity of Lombo-Aortic Radiotherapy for Seminoma. A Review Triggered by an Unusual Case

Abstract

Most patients with testicular seminoma have been treated with a curative intent for decades. Second cancers after radiotherapy for testicular seminoma before 1990 are a growing issue, and are related to previous generation of dose planning and delineating strategies. Among those cancers, hepatocellular carcinoma is an extremely rare occurrence, especially when affecting patients with healthy, noncirrhotic liver. Here, we describe such a case in a patient of our institution, and subsequently review the relevant literature and large epidemiologic studies. Understanding those late and serious toxicity features may help cancer care teams to screen and treat those patients appropriately.

Major improvement in testicular cancer treatment in the second half of the 20th century made it a curable disease, mainly thanks to platinum-based chemotherapy and radiation therapy.^{1, 2} Cancer of the testis has one of the best 5-year survival rate.^{3, 4} Avoiding long-term iatrogenic morbidity is now part of the treatment plan. Among the many malignancies arising from the radiotherapy field of previously treated testicular cancer patients, hepatocellular carcinoma is a rare occurrence. We describe a case of a patient presenting with in-field hepatocellular carcinoma 34 years after radiotherapy for testicular seminoma.

Case Report

A 62-year-old man presented in April 2013 with dehydration and acute renal failure. Relevant medical history reported a left testicular seminoma in 1979, for which he underwent left orchidectomy, adjuvant platinum-based chemotherapy, and radiotherapy. Other history included transient ischemic attack in 2002. He had lost 8 kg over the past 8 months. On examination, there was no symptom of portal hypertension or chronic liver failure. Abdominal ultrasound (US) found a heterogeneous liver mass. Thoracic and abdominal computerized tomography scan (CT scan) found a 16 cm great axis tumor occupying left liver lobe almost entirely, associated with left portal thrombosis (Figure 1). The thrombus was believed to be of tumorous nature. Angio-CT scan of the liver confirmed the invasion of the left portal vein branch by the tumor. The Right liver lobe was not affected and presented normal radiological features. No distant metastases were found. Gastroscopy and colonoscopy showed no suspicious lesions. Alpha-fetoprotein level was 374,000 ng/mL. Other tumor markers were within normal range. Liver biopsies were performed. Pathology report came to the conclusion of a well-differentiated hepatocarcinoma (Figure 2). Liver function test were unaffected. Every major cause of cirrhosis and liver disease was ruled out including alcoholic cirrhosis, autoimmune hepatitis, hemochromatosis, Non-Alcoholic Steato-Hepatitis (NASH) or fatty liver disease, primary biliary cirrhosis, sclerosing cholangitis, and Wilson's disease. Viral serologies were negative for hepatitis B (HBs antigen and HBc antibody), hepatitis C, and cytomegalovirus (CMV), and there were no clinical suspicions of viral hepatitis. No aflatoxin exposure was found. 15 min Indocyanin green (IG) dye retention rate was 9.3% (normal range below 10%) and IG clearance was slightly low at 14.6% (normal range between 18 and 20%). The liver was thus considered healthy and noncirrhotic. Preoperative liver MRI described heterogeneous, T1 hyposignal, and T2 hypersignal lesions occupying the left liver lobe in whole, but confined into it. Coronal enhancement and delayed contrast washout were featured. Of note, right lobe parenchyma appeared homogeneous without any signal anomaly even on early arterial acquisition. After multidisciplinary meeting, the patient was considered fit for surgery. Patient treatment plan consisted in complete surgical resection along with omentectomy and haemostatic splenectomy in June 2013 in a high-volume hepatic surgery center. Pathology report diagnosed a moderately differentiated, stage IV, pT4N0M1 hepatocellular carcinoma with omental metastases and vascular involvement, strongly stained positive for α-fetoprotein⁵ (Figure 3), with a histological grade Edmonson and Steiner II. Myc gene amplification⁶ was not present. Biopsies on adjacent liver showed no signs of hepatic disease (including neither cirrhosis nor hepatocellular carcinoma). There was no histological evidence of hepatic venous outflow tract obstruction, such as periportal or perivenular fibrosis, areas of infarction, and compensatory regenerative nodular hyperplasia. Surgical removal was considered adequate and close follow-up was set up. July 2013 CT scan showed metastatic recurrence in lung and patient was treated with sorafenib, until tumor progression in October 2013. Treatment then consisted in chemotherapy (oxaliplatin, gemcitabin) with an initial partial response followed by a disease progression (liver and lung) after 6 months. Because of ECOG, performance status was still rated 1; a third treatment line with pegylated liposomal doxorubicin along with 5-FU was conducted during 3 months before disease progression. Patient died in October 2014, 18 months after the initial diagnosis. The median survival for stage IV hepatocellular carcinoma is 10.7 months.⁷

Figure 1.

Left hepatic tumor within the radiotherapy field.

Figure 2.

Hepatocarcinoma (a) next to healthy liver parenchyma (b).

Figure 3.

Alpha-fetoprotein stain.

Reviewing patient file, we took a closer look at the initial oncological workup that was carried through in 1979–1980 in order to treat his left testicular seminoma. At that time, patient underwent left orchidectomy, followed by platinum-based chemotherapy and photon-based external beam radiation therapy from January to February 1980. “Hockey stick”, external beam photon radiation therapy of left iliac, and bilateral lombo-aortic lymph nodes were achieved (Figure 4). The dose was 40 Gy with a second radiation field on left renal hilum lymphadenopathy with a 45 Gy dose. In addition, 45 Gy irradiation of bilateral supraclavicular fossa was performed. Many nontarget organs were within the field limits: bladder, small bowel, pancreas, stomach, and left liver. Radiation induced cystitis and proctitis occurred as late toxicity.

Figure 4.

Original 1980 “hockey stick” photon radiation field.

In our diagnostic workup, we noted that the HCC had developed within an area that had been previously irradiated. Our patient had no precancerous lesion of the liver. Knowing that hepatocellular carcinoma occurring on a healthy liver is a rare disease, we hypothesized an association with the external beam radiation therapy on lombo-aortic lymph nodes performed some 30 years before.

Review and Discussion

We looked for other cases of second primary liver cancer occurring after radiation therapy for testicular cancer and found no similar case. The closest paper we found was a series of 3 German patients who developed biliary tract malignancy between 18 and 32 years after being treated with radiotherapy for seminoma or bladder cancer.⁸

Second cancers after radiotherapy treatment are mostly due to environmental and genetic factors. However, in a small subset of patients, those cancers seem related to radiotherapy, as demonstrated by De Gonzalez et al. in a large prospective study using the US Surveillance, Epidemiology and End Results (SEER) program⁹ on 647,672 patients. They found that 8% of second cancers were attributable to radiotherapy. Risk was greater when large fields were used. In-field organs were more likely to be affected. The relative risk of developing a second malignancy increased with longer time from diagnosis. Based on these results, our patient matched those three risk factors and was considered at risk of having second cancer. In fact his radiotherapy course used large lombo-aortic fields that included the left liver, and time from diagnosis of seminoma was prolonged (30 years). He might well have suffered from a hepatocellular carcinoma occurring with no known cause. Nevertheless, we felt necessary to hypothesize that the condition was linked to the previous radiotherapy.

Long-term oncologic morbidity of radiation therapy for testicular cancer was initially a controversial topic and evidence only relied on reports¹⁰ or exploratory studies with poor statistical power.^{11, 12, 13, 14} A recent retrospective review conducted by Hallemeier et al. at the Mayo clinic confirmed excellent outcome in patient with stage I testicular seminoma treated with radiotherapy, even though cardiovascular (12%) and oncologic (19%) late morbidities¹⁵ proved to be not uncommon.

Studies have shown that overall mortality and second cancer-specific mortality among testicular cancer patients who were diagnosed in the 1980 and treated with high-dose radiation therapy were significantly higher than that of first cancer.^{16, 17, 18} On the other hand, the incidence of secondary malignancies among testicular germ cell tumor(TGCT) survivors is higher compared to the general population.¹⁹ However, among all second cancers described, liver tumors are seldom reported.^{8, 20}

It is now known that patients diagnosed in their young age with testicular cancer are at risk of developing a second primary cancer.¹⁹ Genetic factors but also treatment modality like chemotherapy or more specifically here, external beam radiotherapy²¹ appears to be a significant risk factor. With this issue in mind, radiation oncologists tend to use less toxic ways, e.g. proton beam therapy, to treat seminomas.^{22, 23} Although radiotherapy as an adjuvant treatment for low-grade seminoma was considered relatively safe, a significant proportion of patients develop second cancer in the radiotherapy field among other complications.^{20, 24}

In a 2005 study collecting data from more than 40,000 patients, Travis et al. were the first to raise strong evidence-based argument on radiotherapy as an independent risk factor for second malignancy after testicular cancer treatment.²⁵ In 2012, Lewinshtein et al. study confirmed that risk.²⁶ Even though both studies described a significant increase in occurrence of in-field related cancer (lung, pancreas, colon, bladder, kidney), they did not find any liver tumor.

A recent Slovakian study¹⁸ retrospectively analyzed 1367 testicular cancer patients treated between 1970 and 2012 and found significantly higher incidence of second malignancies in that population. They were mostly abdominal or pelvic cancers regardless of treatment for primary testicular cancer, and occurred at a younger age in comparison with the general population.

Using the SEER database on 31,330 patients, Gandaglia et al. asserted that the risk of dying from secondary malignancy significantly increased with advancing stage and age at diagnosis among TGCT survivors.

Using the same data, Schairer et al. demonstrated on a subgroup analysis that mortality from second cancer following testicular cancer was higher for some tumor arising inside the radiotherapy field among testicular cancer patients who were diagnosed during the 1970. These results suggest that treatment of testicular cancer during the 1970 with adjuvant radiotherapy might have weighed adversely on survival from subsequent cancer. However, poor data exist regarding second primary liver cancer, with only 4 cases in the Schairer et al. study.

In 2013, Horwich et al.¹⁷ gathered British data on postradiotherapy outcome for patients treated for stage I seminoma between 1960 and 1992 when adjuvant radiotherapy or adjuvant carboplatin was a standard. They found a significant excess risk of second cancer within the pelvic and abdominal radiation field, contrasting with no such increase organs elsewhere. Again, among 468 second cancer observed, they only elicited 2 liver cancers, with no information specifying whether those liver were healthy or cirrhotic.

With aging population and improving anti-cancer treatments, second primary malignancy in people treated for cancer with radiation therapy is a growing issue.²⁷ This is particularly true for testicular cancer patients. The overall cure rate can be very high.¹⁷ Most patients are young at the time of diagnosis. With right treatment plan, patients can be cured off from cancer and their life expectancy meets general population data. The benefit of long-term screening for this population should be evaluated by further studies.

Conclusion

We reported a case of a man in his 60s treated for hepatocellular carcinoma. Patient had no underlying liver condition. The only oncogenic factor we found was radiotherapy for the treatment of a left testicular seminoma, 30 years earlier. We hypothesized that hepatocellular carcinoma can be a rare but nonetheless serious late-term effect of adjuvant lombo-aortic radiotherapy. We encourage oncological teams to publish similar cases. This report emphasizes the need for proper late-toxicity screening and management in our clinical practice, especially for young patients we intend to cure.

Conflicts of Interest

The authors have none to declare.

Authors’ Contributions

Clement Korenbaum: data and informed consent collecting, literature review, and manuscript writing. Philippe Barthelemy, MD: topic selection and guidance, access to 1980′ radiotherapy file, manuscript supervision, and review. Alina Onea, MD: Pathology pictures and manuscript review. Pierre Salze, MD: radiotherapy expert and manuscript review. Jean-Emmanuel Kurtz, MD PhD: Patient selection and follow-up, manuscript supervision, and review.